Optimizer-Aware Deep Learning for Brain Tumor Classification: A Study Using AlexNet to EfficientNetB0

Optimizer-Aware Deep Learning for Brain Tumor Classification: A Study Using AlexNet to EfficientNetB0 | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Optimizer-Aware Deep Learning for Brain Tumor Classification: A Study Using AlexNet to EfficientNetB0 Hafiz Muhammad Tayyab Khushi, Tehreem Masood, Iftikhar Naseer This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6937303/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract One of the most lethal diseases in the world is brain tumors. Convolutional neural networks (CNNs) and other deep learning-based methods are vital for accurate diagnosis because of their remarkable capacity for learning and prediction. These techniques are widely used in image processing, computer vision, and medical diagnostic tasks such as classification, segmentation, and object detection. In this study, we use the publicly available Figshare brain tumor dataset to compare the performance of different pre-trained deep learning models, including AlexNet, VGG16, VGG19, ResNet50, Xception, InceptionV3, DenseNet121, MobileNetV1, and EfficientNetB0, on four optimizers, such as adaptive moment estimation (Adam), stochastic gradient descent (SGD), root mean square propagation (RMSProp), and Nesterov-accelerated adaptive moment estimation (Nadam). The experimental results show that the VGG19 and EfficientNetB0 models performed exceptionally well with the Adam optimizer, achieving an overall accuracy of 99.13% and a misclassification rate of 0.87%. Additionally, receiver operating characteristic (ROC) curves were calculated, with the EfficientNetB0 model achieving an area under the curve (AUC) value of 100% for each class. It also demonstrated excellent performance on the test images, with a testing accuracy of 99.61%. Brain tumor Figshare transfer learning CNN optimizers MRI Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Introduction Cancer is an abnormal, uncontrolled growth of cells and issues in the body. There are a lot of cancerous diseases in the human body, such as lung cancer, liver cancer, blood cancer, brain tumors, breast cancer, and many more, but as we know, the brain is the most important part of the body, and all the activities inside the body and the movements of the body parts are controlled by the brain. If there is some disturbance in the brain, all other body parts should definitely be in danger. Therefore, the leading disease inside a body is brain tumors around the globe, which can be essential to diagnose, detect, and localize at an early stage to save the patient's life. Brain tumors are broadly classified into two types: primary brain tumors are benign, and secondary brain tumors are malignant [ 1 ]. Most researchers believe benign is not a cancerous disease in the brain, which is some disturbance in the brain that can be cured by some environmental changes, while malignant is a cancerous part of the brain that is too dangerous for the life of humans because of its metastatic nature. Malignant cells can spread to any part of the body from the brain and vice versa [ 2 ]. The World Health Organization (WHO) and the American Brain Tumor Association (ABTA) classify additional malignant tumors into three types: glioma tumors, which come from glial cells and nourish the brain's neurons; meningioma tumors, which come from the meninges and protect the brain and spinal cord's membranes; and pituitary tumors, which come from the pituitary gland at the base of the brain and produce hormones that the body uses for various functions, according to the WHO and the ABTA. Moreover, to analyze the internal anatomy of the brain and to diagnose the tumor in the brain, different medical imaging modalities nowadays have been used, and a few of them are very popular, including X-rays, Computed tomography (CT-Scans), and Magnetic resonance imaging (MRI) based images [ 3 ], X-rays provides the fractured bones details. In contrast, CT scans are used to provide cross-sectional images of the details of body parts from different angles. At the same time, MRI is more beneficial because it provides details regarding soft tissues and cells in the body, and cancer is all about the unusual growth of cells or tissues. In contrast, X-rays and CT scans produce some radiation during the examination of the body parts, which leads to some disturbance in the body with the passage of time. At the same time, MRI is more safer than other imaging modalities. Further, multiple computational approaches have been used to overcome the issue of brain tumor diagnosis, including machine learning, in which different algorithms are applied to an image dataset for the extraction of different features from the images, and based on those features or patterns, either a tumor exists, or a type of abnormality is predicted. However, due to the handcrafted nature of feature extraction and poor performance on large datasets, people move toward deep learning-based approaches, where CNN-based algorithms are applied that extract the best features from the dataset automatically [ 4 ]. The advantages of deep learning over machine learning-based algorithms include their superior capacity to handle large-scale data, better outcomes, and easier application, as well as their superior ability to extract hierarchical features and complicated pattern recognition. Deep learning-based methods are crucial in the medical field for detecting abnormalities in many bodily areas, particularly in the detection of brain tumors, breast cancer, and disorders related to lung cancer. It also helps to find any object detection problem, and it’s all about based on data sampling and their training. To overcome the issue of data sampling and resource consumption, pretrained models have their own importance because they have been trained on a larger dataset and leverage their learning feature ability and knowledge to analyze newly unseen small data samples for solving the current problem. In order to make the most of limited resources and cut down on training time, pre-trained or transfer learning based algorithms are commonly used [ 5 ]. Further, the pretrained models have already been trained on a larger dataset and for a new unseen dataset; common features like color, edges, and textures have already been learned and transferred for the new dataset; and for extracting unique features from the unseen dataset, a few hyperparameters and layers have been added to the pre-trained models for extracting suitable information regarding the problem, which helps to overcome the resource problem; and time savings in terms of training the models from scratch. The main purpose is to diagnose the disturbance in the body or brain, and that transfer learning-based model produced better outputs regarding other approaches that helped radiologists make their decisions on time to secure the lives of patients. The key contributions of this research work are as follows: The main thing that this study adds is that it uses multiple optimizers to validate pre-trained deep learning models and then compares how well they classify brain tumors. To enhance resource management and feature extraction, a transfer learning-based model was employed to assess the performance of these models on the publicly available Figshare brain tumor dataset. A data augmentation approach was applied to capture more generic and specific features from a diverse range of images, addressing the issue of overfitting. Several metrics were used to assess the models' performance. These included precision, sensitivity, specificity, F1-score, accuracy, and intersection over union for misclassification rate. Here is the structure of the research article: The current state-of-the-art approaches are reviewed in Section 2, the material and methodology are presented in Section 3, the results are shown in Section 4, conclusion, and limits and future work is exhibited in Sections 5 and 6, respectively. Literature Review Brain tumor diagnosis utilizing various machine learning and deep learning-based techniques has been the subject of numerous published research articles. Many of them consider the issue of detecting whether tumors exist or not. Many of them focus on the categorization of tumors into their types, and very few of them just focus on the localization of tumor regions in the brain. They have adopted different approaches to diagnose the disturbance in the brain, and a few of them are discussed below. In their study, Khushi et al. [ 6 ] compared various deep learning algorithms with respect to optimizers for brain tumor detection in order to evaluate the model's efficacy using the open-source Br35H dataset. That was the goal of their five-algorithm proposal, which included ResNet50, AlexNet, VGG16, and Stochastic Gradient Descent (SGD), as well as Adam, root mean square propagation (RMSprop), and adaptive moment estimation (AM). In addition, several deep learning models, sensitivity, specificity, accuracy, and MCR (miss classification rate) are utilized as statistical performance evaluation metrics to assess the efficacy of models. Using the SGD optimizer, the AlexNet model outperforms the competition, reaching a total accuracy of 98.79% and a miss classification rate of 1.20%. Another research work was proposed by Khushi et al. [ 5 ], they presented a research methodology for the efficient diagnosis of brain tumors using the proposed EfficientNetB7 model. They also tried different pre-trained models, and out of them, they found better results on the EfficientNetB7 model, so they decided to upgrade that model with the addition of some hyperparameters and layers. The Sartaj brain tumor dataset was obtained from Kaggle for this purpose. The dataset has four classes: glioma, meningioma, pituitary, and no tumor. Prior to the analysis, the images underwent preprocessing that included noise removal and cropping to remove unnecessary parts. In addition, they assessed the suggested model using a variety of performance metrics, such as accuracy and miss classification rate, and ultimately achieved a 98.97% accuracy rate and a 1.02% miss classification rate. Talukder et al. [ 7 ], presented a research work based on brain tumor categorization by different deep learning based pre-trained models using reconstruction and fine-tuning mechanisms. For this, they acquired a publicly available dataset from Figshare, which has three tumor classes and a total of 3064 images. To make the images more visible, they applied a sharp filter before going on to deep learning models. In order to implement the reconstruction principle, they shortened the layers following the activation layer and added an augmentation layer after the models' input layers. The four pre-trained models that were utilized were DenseNet201, Xception, ResNet50V2, and InceptionResNetV2. As compared to the other models, ResNet50V2 outperformed them all with a 99.68% accuracy, 99.49% precision, 99.78% recall, 99.64% F1-Score, and 0.32% MAE (mean absolute error). Another research project was proposed by Khan et al. [ 8 ], also suggested a study to correctly identify brain cancers with the use of a deep learning model. For these purposes, they chose two different datasets: one is Figshare, which is a multiclass-based dataset with 3064 images, and the other is the Harvard Medical dataset, which contains 152 MRI images of two classes. They proposed a 23-layer-based CNN architecture, which caused an overfitting issue, and to overcome this, and they mirrored the planned 23-layer CNN architecture using a pre-trained VGG16 model. On the Figshare dataset, the suggested model attained an accuracy of 97.8% and a precision of 96.5%; on the Harvard medical dataset, it reached an accuracy of 100% with 100% precision, recall, and F1-Scor, respectively. In order to avoid overfitting in future work, it would be beneficial to test the suggested model alongside alternative deep learning models. In their study, Rahman et al. [ 9 ]. demonstrated how to use MRI images and a parallel deep convolutional neural network (PDCNN) model to detect brain cancers. They came up with a PDCNN model that could handle both local and global feature extraction. It did this by using two parallel CNN models on the images, which helped with overfitting. The model also included batch normalization and dropout regularizer layers. For these purposes, they chose three datasets, including the binary tumor classification dataset, the Figshare dataset, and the Sartaj Brain Tumor dataset. For the preprocessing phase, they resized the images with conversion to grayscale images and applied the data augmentation approach to make more data samples. Their proposed model achieved results of 97.33%, 97.60%, and 98.12% accuracy for the binary tumor dataset, the Figshare dataset, and the Sartaj brain tumor dataset, respectively. Using several pre-trained deep learning models for efficient tumor identification, Ullah et al. [ 10 ] reported another study work focused on the categorization of brain tumors. For this, they have selected the publicly accessible Sartaj brain tumor dataset from Kaggle, comprising three tumor classes with one healthy image among four tumor types. Further, they adopted three tumor classes, and before passing to pre-trained models, they applied data augmentation to increase the samples of tumors. In addition, the images were scaled before being fed into deep learning models. The models then underwent modifications to their last three layers, which included a fully connected layer, softmax layers, and classification layers. Among the pre-trained models they utilized, InceptionResNetV2 outperformed the others, achieving an average accuracy of 98.91%, precision of 98.28%, recall of 99.75%, and F1-score of 99%. The other models included InceptionV3, Xception, ResNet18, ResNet50, ResNet101, ShuffleNet, DenseNet201, and MobileNetV2. Even with just two classes, it's important to test the proposed model on both a static and dynamic dataset in order to ensure accuracy. Tummala et al. [ 11 ], presented research for the classification of tumors in the brain with the help of a vision transformer-ensembling approach. For this, they have chosen a well-reputed, publicly available Figshare dataset that has a total of 3064 images of three different tumor classes. For deep learning models, they chose the fine-tuned pre-trained ViT models with variants of B/16, B/32, L/16, and L/32. For result-oriented purposes, the ViT L/32 model achieved the best results, with an accuracy of 98.2% at 384 × 384 image resolution. Further, for ensemble models of all ViT variants, the results achieved 98.7% accuracy at the same image resolution. Further, the ensemble model has to mention the validation accuracy of the models while also finding the other performance evaluation parameters on other brain tumor datasets. A study by Nawaz et al. [ 12 ] examined brain tumors and their accurate identification and classification using the DenseNet41-based CornerNet architecture. Figshare and the Brain MRI dataset were utilized for this objective. Their proposed model comprises three steps: first, they created the annotation to locate the exact region of interest, then further feature extraction was done by CornerNet with DenseNet41 as the backbone, and in the last one-stage detector, CornerNet was applied to locate and classify the tumor in the brain. On both datasets, their model achieved an average result of 98.8% and 98.5% accuracy for the Figshare and Brain MRI datasets, respectively. From a future perspective, the proposed model should be tested on a real-time dataset along with other disease predictions. Maqsood et al. [ 13 ] presented a study for brain tumor detection based on a multimodal deep neural network and a multiclass SVM approach; therefore, they chose two datasets from the open source library Kaggle: one is Brats2018 and the other is Figshare. After applying linear contrast stretching to extract edges, they used a 17-layer customized model for segmentation. For feature extraction, they used a modified MobileNetV2, and for brain tumor classification, they used multiclass SVM and M-SVM. On top of that, the model got 97.47% accuracy on the Brats2018 dataset and 98.92% accuracy on the Figshare dataset. The authors are compelled to utilize an existing model for segmentation and conduct classification using a deep learning model rather than a multiclass support vector machine, hence eliminating the need for seventeen layers. Another research work presented by Malla et al. [ 14 ] for the classification of tumors in the brain with the help of MRI images, was done by applying CNN with a global average pooling layer. For these purposes, they have chosen the Figshare dataset from Kaggle and applied a pre-trained VGG16 model with a global average-pooling layer by freezing model layers to overcome the overfitting issue and banishing the gradient problem. In order to improve the signal-to-noise ratio, eliminate background noise, and maintain edge details, they boost picture contrast during the preprocessing phase. To improve the quality of the features extracted from photos by deep learning models, data augmentation is also used to improve the data samples. A sensitivity level of 98.68%, a specificity level of 99.13%, and a precision level of 99.11% were all obtained by the suggested model. If the author wants to know how well VGG16 performs on datasets, she should utilize other deep learning models; yet, the findings still could be better. Ayadi et al. [ 15 ], presented a research work based on brain tumor classification using a hybrid approach. To achieve this, they analyzed data collected between 2005 and 2010 from two hospitals in China: Nan-Fang Hospital and General Hospital, which is named the Figshare dataset and has records of 233 patients against three tumor classes. The first step is image intensity normalization using min-max techniques. After further processing with DSURF and HOG, the features are input into the SVM classifier, which then sorts the tumors into three categories. A precision of 90.27 percent was attained by the suggested model. In addition, the outcome might be better, and a strategy based on deep learning would be ideal for automated feature extraction and categorization. Polat et al. [ 16 ] published findings from their study on classifying brain tumors using deep learning models trained on magnetic resonance imaging (MR) scans. The Figshare dataset had three types of tumors including glioma, meningioma, and pituitary tumors, and was sourced from the open-source Kaggle library. Further, they passed the dataset to different pre-trained models with different optimizers by adding fully connected layers, and out of all the various deep learning models, the ResNet50 model with the Adadelta optimizer achieved the best result and got 99.02% accuracy. Further testing of deep learning models using various optimizers on the Figshare dataset is necessary to evaluate their performance on this multi-class tumor dataset for future perceptive. Agarwal et al. [ 17 ] presented research work for enhanced brain tumor classification and detection on a publicly available Figshare dataset. For this, they have adopted a two-phase approach, which consisted of enhancing the image contrast by applying ODTWCHE and the classification of the tumor by leaving the pre-trained InceptionV3 model for an accurate refine diagnostic process, and the proposed model achieved an accuracy of 98.89%. Ramakrishnan et al. [ 18 ] proposed a hybrid CNN architecture-based model for brain tumor classification on a publicly available dataset to balance the accuracy and used VGG16, InceptionV3, DenseNet, and ResNet, and their efficiencies were improved by applying oneAPI optimization and achieved overall results with an accuracy of 96.2%. Proposing a deconvolutional residual network, Sekaran et al. [ 19 ] proposed a deformable hierarchical heuristic-based model for 3D brain slice categorization RIDER dataset. The suggested model had an F1-score of 55% and an accuracy of 95% with 80% of precision. Kumar et al. [ 20 ] proposed a deep neural network first using weighted correlation-based feature selection technique and subsequently applied a multivariate neural network to reduce the miss classification rate as well as early brain tumor diagnosis improvements, so addressing the high false alarm rates. The below Table 1 shows the literature review summary of different state-of-the art approaches. Table 1 Literature review survey on different computational approaches regarding brain tumor classification. Ref Techniques Dataset Performance Evaluation Limitations [ 6 ] AlexNet with SGD Br35H Accuracy = 98.79%, MCR = 1.20% Sensitivity = 98.98%, Specificity = 98.58%, F1-Score = 98.82% The author should also have to test other deep learning models for the diagnosis of brain tumors. [ 5 ] Modified EfficientNetB7 Sartaj Accuracy = 98.97%, MCR = 1.02% The proposed approach should be tested on different diseases and real datasets. [ 7 ] ResNet50V2 Figshare Accuracy = 99.68%, Recall = 99.78%, F1-Score = 99.64%, Precision = 99.49%, MAE = 0.32% The proposed model should be used to evaluate different datasets and to diagnose other diseases. [ 8 ] 23-CNN layers with VGG16 Figshare Accuracy = 97.8%, Recall = 96.4%, F1-Score = 96.4%, Precision = 96.5% The proposed model should be tried with other deep learning models to overcome the issue of overfitting. Harvard Medical dataset Accuracy = 100%, Recall = 100%, F1-Score = 100%, Precision = 100%, [ 9 ] PDCNN Binary Accuracy = 97.33%, Recall = 97.50%, F1-Score = 97.50%, Precision = 97.50%, There is still a need to overcome the issue of overfitting, and we should try any existing convolutional neural network model instead of making one from scratch. Figshare Accuracy = 97.60%, Recall = 97.00%, F1-Score = 97.00%, Precision = 97.00%, Sartaj Accuracy = 98.12%, Recall = 97.75%, F1-Score = 98.00%, Precision = 98.00%, [ 10 ] InceptionResNetV2 Sartaj (three tumor classes) Accuracy = 98.91%, Recall = 99.75%, F1-Score = 99%, Precision = 98.28%, There is still a need for improvements; instead of using three classes, there is still a need to validate the whole dataset on the proposed model, as well as on a real-time dataset. [ 11 ] Ensemble of ViTs Figshare Accuracy = 98.70%, Sensitivity = 97.78%, Specificity = 99.42% Further, the ensemble model has to mention the validation accuracy of the models while also finding the other performance evaluation parameters on other brain tumor datasets. [ 12 ] DenseNet41-based CornerNet Framework Figshare Accuracy = 98.8% There should be a need to test the proposed model on a real-time dataset along with other disease predictions Brain MRI Accuracy = 98.5% [ 13 ] Multimodal-M-SVM Brast2018 Accuracy = 97.47%, Specificity = 97.04%, Dice Coef = 96.71%, Sensitivity = 97.22% Instead of using 17 layers, authors have to use an available segmentation model and perform classification using a deep learning model instead of a multiclass SVM. Figshare Accuracy = 98.92%, Specificity = 99.02%, Dice Coef = 99.87%, Sensitivity = 98.82% [ 14 ] VGGNet Figshare Accuracy = 98.93%, Sensitivity = 98.68%, Precision = 99.11%, Specificity = 99.13% The author should use different deep learning models instead of VGG16 to figure out the performance of that model on datasets, and still there is a need for improvements in the results. [ 15 ] DSRUF + HOG Figshare Accuracy = 90.27% There is still a need for improvements in the result, and there should be a need to use a deep learning-based approach for automatic feature extraction and classification. [ 16 ] ResNet50 –Adadelta Figshare Accuracy = 99.02%, AUC = 0.99% There is still a need to check more deep learning models on the Figshare dataset with different optimizers to validate the performance of these multi-class tumor datasets, and more performance evaluation parameters are needed. In the literature review section, we have reviewed multiple research articles regarding brain tumor detection that have been published. Most research articles utilize the publicly accessible Figshare dataset, which comprises three tumor classifications including glioma, meningioma, and pituitary tumor has a record of 233 patients. Many researchers applied smaller, self-created models instead of using a well-defined existing model, and some of them used complex architectures for the detection of tumors in the brain, which further needed some improvements regarding models and results. Materials and Methods Brain tumors become the most lethal disease and the main reason for death in the world because of cancer, and still, there is a need for an accurate diagnosis or detection system that can predict tumors in their early stages to reduce the death ratio. In this research work, we have used different pre-trained models, including AlexNet, VGG16, VGG19, ResNet50, InceptionV3, Xception DenseNet121, MobileNetV1, and EfficientNetB0, with four different optimizers, utilizing Adam, Nadam, RMSprop, and SGD for brain tumor classification, as seen in Fig. 1 . 3.1 - Data Acquisition The overall procedure and workflow of the proposed model are described in Fig. 1 . Further, before applying any computational approach, the first step is to obtain a dataset, which was taken from the open source library Kaggle and named Figshare brain tumor dataset [ 21 ], which has records of 233 patients and was taken from Nan-Fang Hospital and General Hospital, China, from 2005 to 2010. The dataset further has a total of 3064 images categorized into three tumor classes, including glioma, meningioma, and pituitary tumors. Out of 3064 images, the glioma tumor class contains 89 patients with 1426 images, while the meningioma tumor class contains 82 patients with 708 images, and the pituitary tumor class contains 62 patients with 930 images. All images are on three different planes, including axial, coronal, and sagittal, as shown in Fig. 2 . 3.2 - Data Preprocessing To begin, the images were resized to 224 × 224. In order to address the overfitting issues and improve the learning and extraction of features by deep learning models, a data augmentation approach was implemented prior to the dataset being entered into the training and validation phase. This approach generated an increased number of data samples from various perspectives. For data augmentation, a 7% rotation range was applied for rotating the images up to 7 degrees; further width and height shifts were applied to 5% for rotating the images, up to 5% of the total images horizontally and vertically. After that, a zoom range of 10% is applied to zoom in and out of the images by a maximum of 10%, and lastly, randomly flip the images horizontally and vertically. After that, dataset splitting was applied, and the dataset was split with an 85% ratio of training and a 15% ratio of validation, which is shown in Table 2 . Next, before passing data samples to different pre-trained models, label encoding is applied, which is essential for the model to process the labels correctly and to compute the loss during training and evaluation. Table 2 Data distribution of Figshare regarding training and validation w.r.t each class. Tumor Class Total Samples Training Samples Validation Samples Glioma 1426 1212 214 Meningioma 708 602 106 Pituitary 930 790 140 Total 3064 2604 460 3.3 - Pre-trained Deep learning model Additionally, transfer learning-based models are implemented for the purpose of training and validation. As output layers, these models are equipped with the SoftMax classifier for the classification of brain tumors. They consist of a global average pooling layer, a dense layer with ReLu activation function, and a fully connected layer. Pre-trained models are preferred due to their capacity to extract valuable features from images and their prior training on a vast dataset. In contrast to a model that is formed from the ground up, the process of fine-tuning them on a new dataset typically necessitates less training time. When labeled data is scarce or inadequate, transfer learning enables the utilization of knowledge from a related domain that has data. This also helps when we have limited resources, alleviates the issue of overfitting, and enhances the model's capacity to generalize the unobserved data. The first model that we have selected is the AlexNet model, which was introduced by Alex Krizhevsky et al. in 2012 [ 22 ]. There were eight layers in total: five convolutional layers and three fully linked layers. The model also took advantage of the ReLU activation function, which overcomes the vanishing gradient issue, therefore enabling faster network training. To overcome the overfitting problem and cut the spatial dimensions, they also add the dropout and max-pooling layers. The ReLU activation function and dropout layer were added for the first time after the model was trained on 1000 image classes. VGG16 and VGG19 models were also used, constructed in 2014 by the Visual Geometry Group at the University of Oxford. Although the number of convolutional layers varies, the VGG16 and VGG19 models have a similar construction. VGG19 has 19 layers in total, including 16 convolutional layers, whereas VGG16 has 16 levels, 13 of which are convolutional. Both models use stochastic gradient descent (SGD) optimization for training and include three completely connected layers [ 23 ]. The next model was ResNet50 [ 24 ], which was developed by Kaiming He et al. in 2015, and the key contribution of this model is residual learning to solve the vanishing gradient in deep network models. Using residual blocks, this model comprises of two or three convolutional layers with shortcut connections skipping one or more levels. There are a total 50 layers in the ResNet50 model, which is composed of 49 convolutional layers and an output layer that is divided into 16 residual blocks. The Further Xception model [ 25 ], which stands for Extreme Inception, was introduced by Francois Chollet and is based on depthwise separable convolutions. In the Xception model, residual connections are also used, which improves the gradient flow and training accuracy. This model replaces standard convolutions in Inception with depthwise separable convolution followed by pointwise convolution. InceptionV3 [ 26 ], also known as GoogleNet, was introduced by Christian Szegedy et al., which is the third version of the Inception family. This model allows or combines multiple convolutional operations and pooling operations in parallel, allowing the model to capture features at different scales. This model also employs batch normalization to speed up training and uses the RMSProp optimizer. Further, the MobileNetV1[ 27 ] model was selected, which was developed by Andrew G. Howard et al. The primary objective of this model's development was to reduce computational complexity and memory consumption while ensuring high performance. This model used the concepts of depthwise and pointwise separable convolution. There are a total of 13 depthwise separable convolutional layers, each followed by batch normalization, the ReLu6 activation function, and the RMSProp optimizer adopted for the training. Moreover, the DenseNet121 [ 28 ] model was introduced by Gao Huang et al, which belongs to the DenseNet family. DenseNet is comprised of dense blocks, in which each layer is connected to every other layer in a feed-forward manner. This results in a more efficient network and the utilization of features. The quantity of input feature maps placed between the dense blocks is reduced by the use of additional bottleneck layers and transition layers, which downsample the features, and the model is trained on SGD with momentum optimizers. The EfficientNetB0 [ 29 ] model is the baseline model of the EfficientNet family, introduced by Mingxing Tan and Quoc V. Le. The model is built utilizing a compound scaling technique whereby all dimensions of depth, width, and resolution are evenly scaled under control by a set of predetermined scaling factors. The model used mobile inverted bottleneck convolution blocks, which are composites of depthwise separable convolution, squeeze, excitation, and skip connections. Therefore, for assessing the behavior and performance of the deep learning models with the addition of a global average pooling layer, which is responsible for lowering the dimensions of feature maps to a fixed size by taking the average value of each feature map to a fixed size across its spatial dimensions, the above mentioned pre-trained models have been trained on different optimizers, including Adam, SGD, RMSProp, and Nadam. The equations below illustrate how this mechanism operates. $$\:GAP{\left({F}_{in}\right)}_{c}=\:\frac{1}{H\:\times\:\:W}\:{\sum\:}_{i=1}^{H}{\sum\:}_{J=1}^{W}{F}_{in}(i,\:j,\:c)$$ 1 Where \(\:{F}_{in}\) is the feature map that comes as input from the base model with H × W × C, width, height, and channel on which the global average pooling layer is applied for reducing feature dimensions. Further, from the global average pooling layer, an output x is fed into a dense layer with 256 neurons and ReLu activation function to introduce learnable parameters, including biases and weights, which is responsible for enabling models to learn complex features or patterns and relationship from the aggregated features produced by global average pooling layer by also introducing non-linearity by ReLu activation function to learn complex, non-linear mapping from inputs to outputs. $$\:x=GAP{\left({F}_{in}\right)}_{C}$$ 2 $$\:h=ReLu({W}_{1\:}\bullet\:\:x+\:{b}_{1})$$ 3 Where x is the output vector from the global average pooling layer fed into a dense layer with 256 neurons and ReLu activation. \(\:W\) is the weight of a matrix size of 256, and \(\:{b}_{1}\) is the bias vector of size 256, where h is the output from the dense layer which further fed into output layer with SoftMax activation function for the classification. $$\:{y}_{j}=\:\frac{\text{exp}\left({({W}_{2\:\:}\bullet\:\:h\:+\:{b}_{2})}_{j}\right)}{{\sum\:}_{k=1}^{K}\text{e}\text{x}\text{p}\left({({W}_{2\:\:}\bullet\:\:h\:+\:{b}_{2})}_{k}\right)}\:$$ 4 $$\:y=SoftMax({W}_{2\:}\bullet\:\:ReLu\left({W}_{1\:}\bullet\:\:GAP\left({F}_{in}\right)+\:{b}_{1}\right)+\:{b}_{2})$$ 5 Where \(\:{W}_{2\:}\) is the weight matrix of the output layer with size K × 256, while \(\:{b}_{2}\) is the bias vector of output layer with size k. y represents the output of the final layer after applying SoftMax activation function, which shows the probability for each class with size k. All the above-mentioned pre-trained models have been trained on different optimizers, including Adam, SGD, RMSProp, and Nadam, Deep learning or convolutional neural network models depend critically on optimizers since they update the neural network weights and reduce the loss function, therefore enabling the model to learn from the data and hence increase its performance over time. These optimizers also identify the set of parameters including weights and bias that reduce the loss function and track the variations between projected and real output. SGD is the first optimizer we have selected. [ 30 ], which updates the model parameters by taking small steps in the direction of the negative gradient of loss functions. It updates the weights for each training example individually, which can lead to noisy updates but often helps to escape local minima. $$\:{\theta\:}_{t+1}\:=\:{\theta\:}_{t}-\:\eta\:{\nabla\:}_{\theta\:}J\left({\theta\:}_{t}\right)$$ 6 Where: \(\:{\theta\:}_{t}\) are the parameters at time step \(\:t\) , while \(\:\eta\:\) is the learning rate \(\:{\nabla\:}_{\theta\:}J\left({\theta\:}_{t}\right)\) is the gradient of the loss function \(\:J\) with respect to the parameters \(\:\theta\:\) . Subtracting the gradient's multiplication by the learning rate helps to update the parameters and guide them toward a lower loss. The next optimizer that we have chosen is RMSProp [ 31 ], which is an adaptive learning rate method designed for dealing with the issue of vanishing gradients and exploding gradients. It independently modulates the learning rate for every parameter. It first calculates the moving average of squared gradients then adjusts the settings. $$\:{\upsilon\:}_{t}=\:\beta\:{\upsilon\:}_{t-1}+(1-\:\beta\:){\left({\nabla\:}_{\theta\:}J\right({\theta\:}_{t}\left)\:\right)}^{2}$$ 7 $$\:{\theta\:}_{t+1}\:=\:{\theta\:}_{t}-\:\frac{\eta\:}{\sqrt{{\upsilon\:}_{t}+\:ϵ}}\:{\nabla\:}_{\theta\:}J\left({\theta\:}_{t}\right)\:$$ 8 Where: \(\:{\upsilon\:}_{t}\) is the moving average of the squared gradients, \(\:\beta\:\) is the decay rate, which is typically around 0.9, and \(\:ϵ\) is a small constant that is used to prevent division by zero. RMSProp dynamically changes the step size by inversely proportionalizing the learning rate for every parameter to the square root of the total of the recent squared gradient. The next optimizer that we have chosen is Adam [ 32 ], uses momentum in addition to RMSProp. It maintains gradients' uncentered variance (second moment) as well as their mean (first moment). $$\:{m}_{t}=\:{\beta\:}_{1}{m}_{t-1}+(1-\:{\beta\:}_{1}){\nabla\:}_{\theta\:}J\left({\theta\:}_{t}\right)\:$$ 9 $$\:{\upsilon\:}_{t}=\:{\beta\:}_{2}{\upsilon\:}_{t-1}+(1-\:{\beta\:}_{2}){\left({\nabla\:}_{\theta\:}J\right({\theta\:}_{t}\left)\right)}^{2}\:$$ 10 $$\:{\mathcal{m}}_{t}=\:\frac{{m}_{t}}{1-\:{\beta\:}_{1}^{t}}$$ 11 $$\:{\mathcal{v}}_{t}=\:\frac{{\upsilon\:}_{t}}{1-\:{\beta\:}_{2}^{t}}$$ 12 $$\:{\theta\:}_{t+1}=\:{\theta\:}_{t}-\:\frac{\eta\:}{\sqrt{{\mathcal{v}}_{t}}+\:ϵ\:}{\mathcal{m}}_{t}$$ 13 Where: \(\:{m}_{t}\) and \(\:{\upsilon\:}_{t}\) are the first and second-moment estimations, while \(\:{\beta\:}_{1}\) and \(\:{\beta\:}_{2}\) are the decay rates for the moment estimates commonly set to 0.9 and 0.999. Further, \(\:{\mathcal{m}}_{t}\) and \(\:{\mathcal{v}}_{t}\) are the bias-corrected moment estimates, and \(\:ϵ\) is a small constant to prevent division by zero. Adam considers the mean and variance of the gradients to adjust the learning rates for every parameter, therefore enabling handling of both noisy and sparse gradients. The last optimizer that we have chosen is Nadam (Nesterov-accelerated adaptive moment estimation) [ 33 ], which is an extension of Adam that incorporates Nesterov momentum, which provides a more estimate of the gradient by looking ahead in the direction of momentum. Like Adam Optimizer, Nadam updates the biased first and second-moment estimations, computes the bias-corrected first and second-moment estimations, and then updates the parameter by incorporating the Nesterov momentum. $$\:{\theta\:}_{t+1}=\:{\theta\:}_{t}-\:\frac{\eta\:}{\sqrt{{\mathcal{v}}_{t}}+\:ϵ\:}({\beta\:}_{1}{\mathcal{m}}_{t}+\:\frac{(1-\:{\beta\:}_{1}){\nabla\:}_{\theta\:}J\left({\theta\:}_{t}\right)}{1-\:{\beta\:}_{1}^{t}})\:$$ 14 Nadam improves upon Adam by incorporating the Nesterov accelerated gradient, which anticipates the direction of the gradient, leading to faster convergence and better performance. In the above-discussed optimizers, SGD is simple and works well for many problems but can be slow and stuck in local minima sometimes, while RMSProp adjusts the learning rate dynamically and is more suitable for non-stationary objectives. Further, Adam combined the advantages of both momentum and adaptive learning rates, which make him more robust and efficient for wide-ranging problems, and Nadam enhanced his performance by incorporating Nesterov momentum, which often provides better convergence and performance. Further, we have evaluated the different optimizers on different pre-trained models. Every optimizer has its own pros and cons, and their results will be discussed in the above section. Results and Discussions On four different optimizers, including Adam, SGD, RMSProp, and Nadam, this section shows the results of various pre-trained deep learning models, including AlexNet, VGG16, VGG19, ResNet50, Xception, InceptionV3, DenseNet121, MobileNetV1, and EfficientNet B0, on a publicly available dataset, Figshare, for the detection of brain tumors into three categories, glioma, meningioma, and pituitary tumor. Moreover, the performances of the different pre-trained models have been evaluated regarding different statistical performance evaluation parameters, including accuracy [ 34 ], miss classification rate (MCR) [ 6 ], precision also called positive predicted value (PPV) [ 35 ], sensitivity also known as recall or true positive rate (TPR) [ 36 ], specificity also known as true negative rate (TNR) [ 37 ], F1-score [ 38 ], intersection over union (IoU) [ 39 ], negative predictive value (NPV), false negative rate also known as miss rate (FNR), false positive rate also known as fallout (FPR), false discovery rate (FDR), false omission rate (FOR), Matthews correlation coefficient (MCC) [ 40 ], balanced accuracy, and geometric mean [ 41 ]. $$\:Accuracy=\:\frac{TN+TP}{TN+TP+FN+FP}\:\times\:100$$ 15 $$\:Miss\:classification\:rate\:\left(MCR\:\right)=\:\frac{FP+FN}{TN+FN+FP+TP}\:\times\:100$$ 16 $$\:Precision=\:\frac{TP}{TP+FP}\:\times\:100$$ 17 $$\:Sensitivity=\:\frac{TP}{TP+FN}\:\times\:100$$ 18 $$\:Specificity=\:\frac{TN}{TN+FP}\:\times\:100$$ 19 $$\:F1-Score=2\:\times\:\:\frac{Precision\:\times\:Sensitivity}{Precision+Sensitivity}\:\times\:100$$ 20 $$\:Intersection\:over\:Union\:\left(IoU\right)=\:\frac{TP}{TP+FP+FN}\:\times\:100$$ 21 $$\:Negative\:pedictive\:value\:\left(NPV\right)=\:\frac{TN}{TN+FN}\times\:100$$ 22 $$\:False\:negative\:rate\:\left(FNR\right)=\:\frac{FN}{FN+TP}\:\times\:100$$ 23 $$\:False\:positive\:rate\:\left(FPR\right)=\:\frac{FP}{FP+TN}\:\times\:100$$ 24 $$\:False\:discovery\:rate\:\left(FDR\right)=\:\frac{FP}{FP+TP}\:\times\:100$$ 25 $$\:False\:omission\:rate\:\left(FOR\right)=\:\frac{FN}{FN+TN}\:\times\:100$$ 26 $$\:Matthews\:correlation\:coefficient\:\left(MCC\right)=\:\frac{\left(TP\:\times\:TN\right)-(FP\:\times\:FN)}{\sqrt{(TP+FP)(TP+FN)(TN+FP)(TN+FN)}}\:\times\:100$$ 27 $$\:Balance\:accuracy=\:\frac{Sensitivity+Specificity}{2}\:\times\:100$$ 28 $$\:Geometric\:mean=\:\sqrt{Sensitivity\:\times\:Specificity}\:\times\:100$$ 29 TP, or the true positive value, shown in the preceding equations indicates how precisely the model predicts positive classes. TN, the actual negative value, demonstrates how well the model forecasts the negative class. Furthermore, FN stands for the false negative value, which shows the incorrectly projected negative class by the model; FP stands for the false positive, which shows the improperly expected positive class by the model. Based on the above-mentioned values, all the aforementioned statistical performance evaluation criteria were computed; where accuracy is the proportion of correctly identified cases among the total instances. The miss classification rate indicates among all the cases the improperly classified ones. While sensitivity is the proportion of actual positives accurately detected by the model, precision is the proportion of true positive forecasts among all positive predictions. Specificity is the model's proper identification of the actual negatives. The F1-score is the harmonic mean of sensitivity that is, recall and accuracy. The Jaccard index, sometimes known as the IoU, shows how often expected and actual classes overlap. Moreover, NPV shows among all negative forecasts the fraction of real negative predictions. FNR reveals the percentage of true positives mistakenly projected as negatives. While FDR displays the proportion of false positive predictions among all positive predictions, FPR reveals the fraction of real negatives that are mistakenly recognized as positives. FOR shows among all negative predictions the percentage of false negative predictions. Moreover, MCC shows the relationship between expected and actual classes; balance accuracy offers the average of sensitivity and specificity, which is more helpful for an imbalanced dataset; and the geometric mean also shows the geometric mean of sensitivity and specificity, which is likewise helpful for an imbalanced dataset. Table 3 shows the performance evaluation of the AlexNet model regarding different performance evaluation parameters, including accuracy, MCR, precision, sensitivity, specificity, F1-score, and IoU of each class on different optimizers, while Table 4 shows the overall performance of the AlexNet model regarding the accuracy, MCR, precision, sensitivity, specificity, F1-score, and IoU, and Table 5 shows the overall performance of the AlexNet model regarding NPV, FNR, FPR, FDR, FOR, MCC, balance accuracy (BA), and geometric mean (GM) on different optimizers. Table 3 Performance of AlexNet model regarding each class on different optimizers. Optimizers Image Class Accuracy MCR Precision Sensitivity Specificity F1-Score IoU Adam Glioma 95.00% 5.00% 91.70% 98.13% 92.28% 94.81% 90.13% Meningioma 94.78% 5.22% 93.62% 83.02% 98.31% 88.00% 78.57% Pituitary 98.04% 1.96% 97.81% 95.71% 99.06% 96.75% 93.71% SGD Glioma 83.26% 16.74% 84.42% 78.50% 87.40% 81.36% 68.57% Meningioma 82.83% 17.17% 63.92% 58.49% 90.11% 61.08% 43.97% Pituitary 94.35% 5.65% 84.76% 99.29% 92.19% 91.45% 84.24% RMSProp Glioma 94.13% 5.87% 96.06% 91.12% 96.75% 93.53% 87.84% Meningioma 91.96% 8.04% 89.66% 73.58% 97.46% 80.83% 67.83% Pituitary 93.48% 6.52% 82.35% 100.00% 90.63% 90.32% 82.35% Nadam Glioma 93.70% 6.30% 92.63% 93.93% 93.50% 93.27% 87.39% Meningioma 90.43% 9.57% 89.74% 66.04% 97.74% 76.09% 61.40% Pituitary 94.57% 5.43% 84.85% 100.00% 92.19% 91.80% 84.85% Table 3 shows the comparison of different evaluation parameters on different optimizers of the AlexNet model regarding each class. The model achieved the best results on Adam Optimizer, with an accuracy of 95%, 94.78%, and 98.04% for glioma, meningioma, and pituitary tumor classes, respectively, with MCRs of 5.0%, 5.22%, and 1.96% for each class, respectively. While the model achieved the lowest results on the SGD optimizer, achieving an accuracy of 83.26%, 82.83%, and 94.35% for glioma, meningioma, and pituitary tumor classes, respectively, with an MCR of 16.74%, 17.17%, and 5.65% for each class respectively. Table 4 Overall performance of AlexNet model on different optimizers (Continue). Optimizers Accuracy MCR Precision Sensitivity Specificity F1-Score IoU Adam 95.94% 4.06% 94.38% 92.29% 96.55% 93.19% 87.47% SGD 86.81% 13.19% 77.70% 78.76% 89.90% 77.96% 65.60% RMSProp 93.19% 6.81% 89.36% 88.24% 94.94% 88.23% 79.34% Nadam 92.90% 7.10% 89.07% 86.65% 94.47% 87.05% 77.88% Table 4 shows the overall performance of the AlexNet model on different optimizers regarding accuracy, MCR, precision, sensitivity, specificity, F1-score, and IoU. On all optimizers, Adam Optimizer produced the best results, and the model achieved results such as accuracy of 95.94%, MCR of 4.06%, precision of 94.38%, sensitivity of 92.29%, specificity of 96.55%, F1-score of 93.19%, and IoU of 87.47%. The SGD optimizer produced the worst results, and the model achieved an accuracy of 86.81% with an MCR of 13.19%, precision of 77.70%, sensitivity of 78.76%, specificity of 89.90%, F1-score of 77.96%, and IoU of 65.60%. While RMSProp and Nadam optimizers produced moderate results better than SGD, and the model achieved results on RMSProp with an accuracy of 93.19%, MCR of 6.81%, precision of 89.36%, sensitivity of 88.24%, specificity of 94.94%, F1-score of 88.23%, and IoU of 79.34%. AlexNet achieved results on Nadam Optimizer with an accuracy of 92.90%, MCR of 7.10%, precision of 89.07%, sensitivity of 86.65%, specificity of 94.47%, F1-score of 87.05%, and IoU of 77.88%. Table 5 Overall performance of the AlexNet model on different optimizers Optimizers NPV FNR FPR FDR FOR MCC BA GM Adam 97.16% 7.71% 3.45% 5.62% 2.84% 89.20% 94.42% 94.29% SGD 89.97% 21.24% 10.10% 22.30% 10.03% 69.03% 84.33% 83.70% RMSProp 95.03% 11.76% 5.06% 10.64% 4.97% 83.46% 91.59% 91.26% Nadam 95.08% 13.35% 5.53% 10.93% 4.92% 81.79% 90.56% 90.02% Table 5 presents the overall performance of the AlexNet model in terms of NPV, FNR, FDR, FOR, MCC, BA, and GM across different optimizers. As shown, the Adam optimizer produced the best results, while RMSProp and Nadam delivered moderate results, and the SGD optimizer performed the worst. According to the performance metrics in Table 4 , the model achieved the following values with the Adam optimizer: NPV of 97.16%, FNR of 7.71%, FPR of 3.45%, FDR of 5.62%, FOR of 2.84%, MCC of 89.02%, BA of 94.42%, and GM of 94.29%. In contrast, the model performed less effectively with the SGD optimizer, achieving an NPV of 89.97%, FNR of 21.24%, FPR of 10.10%, FDR of 22.30%, FOR of 10.03%, MCC of 69.03%, BA of 84.33%, and GM of 83.70%. With the RMSProp optimizer, the model attained an NPV of 95.03%, FNR of 11.76%, FPR of 5.06%, FDR of 10.64%, FOR of 4.97%, MCC of 83.46%, and BA and GM values of 91.59% and 91.26%, respectively. Meanwhile, with the Nadam optimizer, the model achieved an NPV of 95.08%, FNR of 13.35%, FPR of 5.53%, FDR of 10.93%, FOR of 4.92%, MCC of 81.79%, and BA and GM values of 90.56% and 90.02%, respectively. Additionally, Table 6 presents the performance evaluation of the VGG16 model across various performance metrics, including accuracy, misclassification rate (MCR), precision, sensitivity, specificity, F1-score, and IoU for each class using different optimizers. Table 7 displays the overall performance of the VGG16 model in terms of accuracy, MCR, precision, sensitivity, specificity, F1-score, and IoU, while Table 8 summarizes the overall performance of the VGG16 model based on NPV, FNR, FPR, FDR, FOR, MCC, BA, and GM with different optimizers. Table 6 Performance of VGG16 model regarding each class on different optimizers. Optimizers Image Class Accuracy MCR Precision Sensitivity Specificity F1-Score IoU Adam Glioma 98.04% 1.96% 97.24% 98.60% 97.56% 97.91% 95.91% Meningioma 97.39% 2.61% 96.08% 92.45% 98.87% 94.23% 89.09% Pituitary 98.91% 1.09% 97.87% 98.57% 99.06% 98.22% 96.50% SGD Glioma 95.22% 4.78% 94.44% 95.33% 95.12% 94.88% 90.27% Meningioma 92.61% 7.39% 86.00% 81.13% 96.05% 83.50% 71.67% Pituitary 97.39% 2.61% 94.44% 97.14% 97.50% 95.77% 91.89% RMSProp Glioma 98.04% 1.96% 97.67% 98.13% 97.97% 97.90% 95.89% Meningioma 97.39% 2.61% 97.96% 90.57% 99.44% 94.12% 88.89% Pituitary 98.04% 1.96% 94.56% 99.29% 97.50% 96.86% 93.92% Nadam Glioma 98.70% 1.30% 99.52% 97.66% 99.59% 98.58% 97.21% Meningioma 97.83% 2.17% 93.64% 97.17% 98.02% 95.37% 91.15% Pituitary 98.70% 1.30% 97.86% 97.86% 99.06% 97.86% 95.80% Table 6 shows the comparison of different evaluation parameters on different optimizers of the VGG16 model regarding each class. The model achieved the best results on Nadam Optimizer, with an accuracy of 98.70%, 97.38%, and 98.70% for glioma, meningioma, and pituitary tumor classes, respectively, with MCRs of 1.30%, 2.17%, and 1.30% for each class, respectively. While the model achieved the lowest results on the SGD optimizer, achieving an accuracy of 95.22%, 92.61%, and 97.39% for glioma, meningioma, and pituitary tumor classes, respectively, with an MCR of 4.78%, 7.39%, and 2.61% for each class respectively. Table 7 Overall performance of VGG16 model on different optimizers (Continue). Optimizers Accuracy MCR Precision Sensitivity Specificity F1-Score IoU Adam 98.12% 1.88% 97.06% 96.54% 98.50% 96.79% 93.83% SGD 95.07% 4.93% 91.63% 91.20% 96.22% 91.38% 84.61% RMSProp 97.83% 2.17% 96.73% 95.99% 98.30% 96.29% 92.90% Nadam 98.41% 1.59% 97.01% 97.56% 98.89% 97.27% 94.72% Table 7 presents the overall performance of the VGG16 model using different optimizers in terms of accuracy, MCR, precision, sensitivity, specificity, F1-score, and IoU. Among all optimizers, the Nadam optimizer produced the best results, with the model achieving an accuracy of 98.41%, an MCR of 1.59%, a precision of 97.01%, a sensitivity of 97.56%, a specificity of 98.89%, an F1-score of 97.27%, and an IoU of 94.72%. In contrast, the SGD optimizer yielded the poorest performance, with the model achieving an accuracy of 95.07%, an MCR of 4.93%, a precision of 91.63%, a sensitivity of 91.20%, a specificity of 96.22%, an F1-score of 91.38%, and an IoU of 84.61%. The RMSProp and Adam optimizers produced moderate results, better than SGD. Using RMSProp, the model achieved an accuracy of 97.83%, an MCR of 2.17%, a precision of 96.73%, a sensitivity of 95.99%, a specificity of 98.30%, an F1-score of 96.29%, and an IoU of 92.90%. With the Adam optimizer, VGG16 achieved an accuracy of 98.12%, an MCR of 1.88%, a precision of 97.06%, a sensitivity of 96.54%, a specificity of 98.50%, an F1-score of 96.79%, and an IoU of 93.83%. Table 8 Overall performance of VGG16 model on different optimizers Optimizers NPV FNR FPR FDR FOR MCC BA GM Adam 98.63% 3.46% 1.50% 2.94% 1.37% 95.12% 97.52% 97.50% SGD 96.36% 8.80% 3.78% 8.37% 3.64% 87.53% 93.71% 93.61% RMSProp 98.43% 4.01% 1.70% 3.27% 1.57% 94.39% 97.15% 97.11% Nadam 98.74% 2.44% 1.11% 2.99% 1.26% 96.38% 98.23% 98.23% Table 8 presents the overall performance of the VGG16 model in terms of NPV, FNR, FDR, FOR, MCC, BA, and GM across different optimizers. As observed, the Nadam optimizer produced the best results, while RMSProp and Adam yielded moderate results, and the SGD optimizer produced the lowest performance. Based on the performance parameters from Table 6 , the model achieved the following values using the Adam optimizer: NPV of 98.63%, FNR of 3.46%, FPR of 1.50%, FDR of 2.94%, FOR of 1.37%, MCC of 95.12%, BA of 97.52%, and GM of 97.50%. Conversely, with the SGD optimizer, the model showed reduced performance with an NPV of 96.36%, FNR of 8.80%, FPR of 3.78%, FDR of 8.37%, FOR of 3.64%, MCC of 87.53%, BA of 93.71%, and GM of 93.61%. Additionally, with the RMSProp optimizer, the model achieved an NPV of 98.43%, FNR of 4.01%, FPR of 1.70%, FDR of 3.27%, FOR of 1.57%, MCC of 94.39%, and BA and GM of 97.15% and 97.11%, respectively. The Nadam optimizer delivered the best results with an NPV of 98.74%, FNR of 2.44%, FPR of 1.11%, FDR of 2.99%, FOR of 1.26%, MCC of 96.38%, and BA and GM of 98.23% and 98.23%, respectively. Table 9 provides the performance evaluation of the VGG19 model based on various parameters, including accuracy, MCR, precision, sensitivity, specificity, F1-score, and IoU for each class across different optimizers. Table 10 summarizes the overall performance of the VGG19 model regarding accuracy, MCR, precision, sensitivity, specificity, F1-score, and IoU. Finally, Table 11 presents the overall performance of the VGG19 model in terms of NPV, FNR, FPR, FDR, FOR, MCC, balanced accuracy (BA), and geometric mean (GM) across different optimizers. Table 9 Performance of VGG19 model regarding each class on different optimizers. Optimizers Image Class Accuracy MCR Precision Sensitivity Specificity F1-Score I0U Adam Glioma 99.13% 0.87% 99.07% 99.07% 99.19% 99.07% 98.15% Meningioma 98.70% 1.30% 98.08% 96.23% 99.44% 97.14% 94.44% Pituitary 99.57% 0.43% 98.59% 100.00% 99.38% 99.29% 98.59% SGD Glioma 95.22% 4.78% 93.24% 96.73% 93.90% 94.95% 90.39% Meningioma 92.61% 7.39% 91.86% 74.53% 98.02% 82.29% 69.91% Pituitary 96.52% 3.48% 90.79% 98.57% 95.63% 94.52% 89.61% RMSProp Glioma 98.91% 1.09% 99.06% 98.60% 99.19% 98.83% 97.69% Meningioma 97.83% 2.17% 97.06% 93.40% 99.15% 95.19% 90.83% Pituitary 98.91% 1.09% 96.55% 100.00% 98.44% 98.25% 96.55% Nadam Glioma 98.26% 1.74% 98.13% 98.13% 98.37% 98.13% 96.33% Meningioma 98.04% 1.96% 96.19% 95.28% 98.87% 95.73% 91.82% Pituitary 99.35% 0.65% 98.58% 99.29% 99.38% 98.93% 97.89% Table 9 shows the comparison of different evaluation parameters on different optimizers of the VGG19 model regarding each class. The model achieved the best results on Adam Optimizer, with an accuracy of 99.13%, 98.70%, and 99.57% for glioma, meningioma, and pituitary tumor classes, respectively, with MCRs of 0.87%, 1.30%, and 0.43% for each class, respectively. While the model achieved the lowest results on the SGD optimizer, achieving accuracy of 95.22%, 92.61%, and 96.52% for glioma, meningioma, and pituitary tumor classes, respectively, with MCRs of 4.78%, 7.39%, and 3.48% for each class respectively. Table 10 Overall performance of VGG19 model on different optimizers (Continue). Optimizers Accuracy MCR Precision Sensitivity Specificity F1-Score I0U Adam 99.13% 0.87% 98.58% 98.43% 99.33% 98.50% 97.06% SGD 94.78% 5.22% 91.96% 89.94% 95.85% 90.59% 83.30% RMSProp 98.55% 1.45% 97.56% 97.33% 98.93% 97.42% 95.02% Nadam 98.55% 1.45% 97.63% 97.57% 98.87% 97.60% 95.35% Table 10 shows the overall performance of the VGG19 model on different optimizers regarding accuracy, MCR, precision, sensitivity, specificity, F1-score, and IoU. On all optimizers, Adam Optimizer produced the best results, and the model achieved results such as accuracy of 99.13%, MCR of 0.87%, precision of 98.58%, sensitivity of 98.43%, specificity of 99.33%, F1-score of 98.50%, and IoU of 97.06%. The SGD optimizer produced the worst results, and the model achieved an accuracy of 94.78% with an MCR of 5.22%, precision of 91.96%, sensitivity of 89.94%, specificity of 95.85%, F1-score of 90.59%, and IoU of 83.30%. While RMSProp and Nadam optimizers produced moderate results better than SGD, and the model achieved results on RMSProp with an accuracy of 98.55%, MCR of 1.45%, precision of 97.56%, sensitivity of 97.33%, specificity of 98.93%, F1-score of 97.42%, and IoU of 95.02%. while VGG19 achieved results on Nadam Optimizer with an accuracy of 98.55%, MCR of 1.45%, precision of 97.63%, sensitivity of 97.57%, specificity of 98.87%, F1-score of 97.60%, and IoU of 95.35%. Table 11 Overall performance of the VGG19 model on different optimizers. Optimizers NPV FNR FPR FDR FOR MCC BA GM Adam 99.35% 1.57% 0.67% 1.42% 0.65% 97.78% 98.88% 98.88% SGD 96.40% 10.06% 4.15% 8.04% 3.60% 86.25% 92.90% 92.62% RMSProp 98.94% 2.67% 1.07% 2.44% 1.06% 96.28% 98.13% 98.11% Nadam 98.88% 2.43% 1.13% 2.37% 1.12% 96.45% 98.22% 98.21% Table 11 shows the overall performance of the VGG19 model regarding NPV, FNR, FDR, FOR, MCC, BA, and GM on different optimizers. As we see above, Adam Optimizer produced better results, RMSProp and Nadam produced moderate results, and SGD Optimizer produced the worst results. Regarding Table 9 performance parameters, the model achieved values on Adam Optimizer as NPV of 99.35%, FNR of 1.57%, FPR of 0.67%, FDR of 1.42%, FOR of 0.65%, MCC of 97.78%, BA of 98.88%, and GM of 98.88%. While on the SGD optimizer, the model achieved less results, as NPV of 96.40%, FNR of 10.06%, FPR of 4.15%, FDR of 8.04%, FOR of 3.60%, an MCC of 86.25%, a BA of 92.90%, and a GM of 92.62%. Further, the RMSProp model achieved a NPV of 98.94%, a FNR of 2.67%, FPR of 1.07%, FDR of 2.44%, FOR of 1.06%, MCC of 96.28%, BA and GM of 98.13%, and 98.11%, respectively. While on Nadam Optimizer, the model achieved the best results, as NPV of 98.88%, FNR of 2.43%, FPR of 1.13%, FDR of 2.37%, FOR of 1.12%, MCC of 96.45%, BA and GM of 98.22% and 98.21%, respectively. Additionally, Table 12 shows the performance evaluation of the ResNet50 model regarding different performance evaluation parameters, including accuracy, MCR, precision, sensitivity, specificity, F1-score, and IoU of each class on different optimizers, while Table 13 shows the overall performance of the ResNet50 model regarding the accuracy, MCR, precision, sensitivity, specificity, F1-score, and IoU, and Table 14 shows the overall performance of the ResNet50 model regarding NPV, FNR, FPR, FDR, FOR, MCC, BA, and GM on different optimizers. Table 12 Performance of ResNet50 model regarding each class on different optimizers. Optimizers Image Class Accuracy MCR Precision Sensitivity Specificity F1-Score IoU Adam Glioma 98.91% 1.09% 99.06% 98.60% 99.19% 98.83% 97.69% Meningioma 97.61% 2.39% 97.03% 92.45% 99.15% 94.69% 89.91% Pituitary 98.70% 1.30% 95.89% 100.00% 98.13% 97.90% 95.89% SGD Glioma 91.96% 8.04% 90.41% 92.52% 91.46% 91.45% 84.26% Meningioma 87.17% 12.83% 78.31% 61.32% 94.92% 68.78% 52.42% Pituitary 93.48% 6.52% 84.81% 95.71% 92.50% 89.93% 81.71% RMSProp Glioma 99.13% 0.87% 99.07% 99.07% 99.19% 99.07% 98.15% Meningioma 97.61% 2.39% 98.97% 90.57% 99.72% 94.58% 89.72% Pituitary 98.04% 1.96% 93.96% 100.00% 97.19% 96.89% 93.96% Nadam Glioma 99.13% 0.87% 98.61% 99.53% 98.78% 99.07% 98.16% Meningioma 96.30% 3.70% 98.90% 84.91% 99.72% 91.37% 84.11% Pituitary 97.17% 2.83% 91.50% 100.00% 95.94% 95.56% 91.50% Table 12 shows the comparison of different evaluation parameters on different optimizers of the ResNet50 model regarding each class. The model achieved the best results on Adam Optimizer, with an accuracy of 98.91%, 97.61%, and 98.70% for glioma, meningioma, and pituitary tumor classes, respectively, with MCRs of 1.09%, 2.39%, and 1.30% for each class, respectively. While the model achieved the lowest results on the SGD optimizer, achieving an accuracy of 91.96%, 87.17%, and 93.48% for glioma, meningioma, and pituitary tumor classes, respectively, with MCRs of 8.04%, 12.83%, and 6.52% for each class respectively. Table 13 Overall performance of ResNet50 model on different optimizers (Continue). Optimizers Accuracy MCR Precision Sensitivity Specificity F1-Score IoU Adam 98.41% 1.59% 97.33% 97.02% 98.82% 97.14% 94.49% SGD 90.87% 9.13% 84.51% 83.19% 92.96% 83.39% 72.79% RMSProp 98.26% 1.74% 97.33% 96.54% 98.70% 96.84% 93.94% Nadam 97.54% 2.46% 96.34% 94.81% 98.15% 95.33% 91.26% Table 13 shows the overall performance of the ResNet50 model on different optimizers regarding accuracy, MCR, precision, sensitivity, specificity, F1-score, and IoU. On all optimizers, Adam Optimizer produced the best results, and the model achieved results such as accuracy of 98.41%, MCR of 1.59%, precision of 97.33%, sensitivity of 97.02%, specificity of 98.82%, F1-score of 97.14%, and IoU of 94.49%. The SGD optimizer produced the worst results, and the model achieved an accuracy of 90.87% with MCR of 9.13%, precision of 84.51%, sensitivity of 83.19%, specificity of 92.96%, F1-score of 83.39%, and IoU of 72.79%. While RMSProp and Nadam optimizers produced moderate results better than SGD, and the model achieved results on RMSProp as accuracy of 98.26%, MCR of 1.74%, precision of 97.33%, sensitivity of 96.54%, specificity of 98.70%, F1-score of 96.84%, and IoU of 93.94%. while ResNet50 achieved results on Nadam Optimizer as accuracy of 97.54%, MCR of 2.46%, precision of 96.34%, sensitivity of 94.81%, specificity of 98.15%, F1-score of 95.33%, and IoU of 91.26%. Table 14 Overall performance of the ResNet50 model on different optimizers Optimizers NPV FNR FPR FDR FOR MCC BA GM Adam 98.85% 2.98% 1.18% 2.67% 1.15% 95.87% 97.92% 97.90% SGD 93.50% 16.81% 7.04% 15.49% 6.50% 76.72% 88.07% 87.46% RMSProp 98.81% 3.46% 1.30% 2.67% 1.19% 95.33% 97.62% 97.58% Nadam 98.42% 5.19% 1.85% 3.66% 1.58% 93.18% 96.48% 96.37% Table 14 shows the overall performance of the ResNet50 model regarding NPV, FNR, FDR, FOR, MCC, BA, and GM on different optimizers. As we see above, Adam Optimizer produced better results, RMSProp and Nadam produced moderate results, and SGD Optimizer produced the worst results. Regarding Table 12 performance parameters, the model achieved values on Adam Optimizer as NPV of 98.85%, FNR of 2.98%, FPR of 1.18%, FDR of 2.67%, FOR of 1.15%, MCC of 95.87%, BA of 97.92%, and GM of 97.90%. While on the SGD optimizer, the model achieved less results, as NPV of 93.50%, FNR of 16.81%, FPR of 7.04%, FDR of 15.49%, FOR of 6.50%, a MCC of 76.72%, a BA of 88.07%, and a GM of 87.46%. Further, the RMSProp model achieved a NPV of 98.81%, a FNR of 3.46%, FPR of 1.30%, FDR of 2.67%, FOR of 1.19%, MCC of 95.33%, BA and GM of 97.62%, and 97.58%, respectively. While on Nadam Optimizer, the model achieved results as NPV of 98.42%, FNR of 5.19%, FPR of 1.85%, FDR of 3.66%, FOR of 1.58%, MCC of 93.18%, BA and GM of 96.48% and 96.37%, respectively. Moreover, Table 15 shows the performance evaluation of the Xception model regarding different performance evaluation parameters, including accuracy, MCR, precision, sensitivity, specificity, F1-score, and IoU of each class on different optimizers, while Table 16 shows the overall performance of the Xception model regarding accuracy, MCR, precision, sensitivity, specificity, F1-score, and IoU, and Table 17 shows the overall performance of the Xception model regarding NPV, FNR, FPR, FDR, FOR, MCC, BA, and GM on different optimizers. Table 15 Performance of the Xception model regarding each class on different optimizers. Optimizers Image Class Accuracy MCR Precision Sensitivity Specificity F1-Score IoU Adam Glioma 99.35% 0.65% 100.00% 98.60% 100.00% 99.29% 98.60% Meningioma 97.83% 2.17% 96.15% 94.34% 98.87% 95.24% 90.91% Pituitary 98.48% 1.52% 95.86% 99.29% 98.13% 97.54% 95.21% SGD Glioma 56.74% 43.26% 51.86% 97.66% 21.14% 67.75% 51.23% Meningioma 77.39% 22.61% 66.67% 3.77% 99.44% 7.14% 3.70% Pituitary 74.13% 25.87% 70.59% 25.71% 95.31% 37.70% 23.23% RMSProp Glioma 99.78% 0.22% 100.00% 99.53% 100.00% 99.77% 99.53% Meningioma 98.48% 1.52% 98.06% 95.28% 99.44% 96.65% 93.52% Pituitary 98.70% 1.30% 96.53% 99.29% 98.44% 97.89% 95.86% Nadam Glioma 99.35% 0.65% 99.53% 99.07% 99.59% 99.30% 98.60% Meningioma 97.83% 2.17% 97.06% 93.40% 99.15% 95.19% 90.83% Pituitary 98.48% 1.52% 95.86% 99.29% 98.13% 97.54% 95.21% Table 15 shows the comparison of different evaluation parameters on different optimizers of the Xception model regarding each class. The model achieved the best results on RMSProp Optimizer, with an accuracy of 99.78%, 98.48%, and 98.70% for glioma, meningioma, and pituitary tumor classes, respectively, with MCRs of 0.22%, 1.52%, and 1.30% for each class, respectively. While the model achieved the lowest results on the SGD optimizer, achieving an accuracy of 56.74%, 77.39%, and 74.13% for glioma, meningioma, and pituitary tumor classes, respectively, with MCRs of 43.26%, 22.61%, and 25.87% for each class respectively. Table 16 Overall performance of the Xception model on different optimizers (Continue). Optimizers Accuracy MCR Precision Sensitivity Specificity F1-Score IoU Adam 98.55% 1.45% 97.34% 97.41% 99.00% 97.36% 94.90% SGD 69.42% 30.58% 63.04% 42.38% 71.96% 37.53% 26.05% RMSProp 98.99% 1.01% 98.20% 98.03% 99.29% 98.10% 96.30% Nadam 98.55% 1.45% 97.48% 97.25% 98.96% 97.34% 94.88% Table 16 shows the overall performance of the Xception model on different optimizers regarding accuracy, MCR, precision, sensitivity, specificity, F1-score, and IoU. On all optimizers, RMSProp Optimizer produced the best results, and the model achieved results such as accuracy of 98.99%, MCR of 1.01%, precision of 98.20%, sensitivity of 98.03%, specificity of 99.29%, F1-score of 98.10%, and IoU of 96.30%. The SGD optimizer produced the worst results, and the model achieved an accuracy of 69.42% with MCR of 30.58%, precision of 63.04%, sensitivity of 42.38%, specificity of 71.96%, F1-score of 37.53%, and IoU of 26.05%. While Adam and Nadam optimizers produced moderate results better than SGD, and model achieved results on Adam as accuracy of 98.55%, MCR of 1.45%, precision of 97.34%, sensitivity of 97.41%, specificity of 99.00%, F1-score of 97.36%, and IoU of 94.90%. While Xception achieved results on Nadam Optimizer as accuracy of 98.55%, MCR of 1.45%, precision of 97.48%, sensitivity of 97.25%, specificity of 98.96%, F1-score of 97.34%, and IoU of 94.88%. Table 17 Overall performance of the Xception model on different optimizers. Optimizers NPV FNR FPR FDR FOR MCC BA GM Adam 98.93% 2.59% 1.00% 2.66% 1.07% 96.38% 98.20% 98.19% SGD 81.11% 57.62% 28.04% 36.96% 18.89% 20.16% 57.17% 38.10% RMSProp 99.29% 1.97% 0.71% 1.80% 0.71% 97.33% 98.66% 98.65% Nadam 98.97% 2.75% 1.04% 2.52% 1.03% 96.22% 98.10% 98.09% Table 17 shows the overall performance of the Xception model regarding NPV, FNR, FDR, FOR, MCC, BA, and GM on different optimizers. As we see above, RMSProp Optimizer produced better results, Adam and Nadam produced moderate results, and SGD Optimizer produced the worst results. Regarding Table 15 performance parameters, the model achieved values on Adam Optimizer as NPV of 98.93%, FNR of 2.59%, FPR of 1.00%, FDR of 2.66%, FOR of 1.07%, MCC of 96.38%, BA of 98.20%, and GM of 98.19%. While on the SGD optimizer, the model achieved less results, as NPV of 81.11%, FNR of 57.62%, FPR of 28.04%, FDR of 36.96%, FOR of 18.89%, a MCC of 20.16%, a BA of 57.17%, and a GM of 38.10%. Further, the RMSProp model achieved a NPV of 99.29%, a FNR of 1.97%, FPR of 0.71%, FDR of 1.80%, FOR of 0.71%, MCC of 97.33%, BA and GM of 98.66%, and 98.65%, respectively. While on Nadam Optimizer, the model achieved results as NPV of 98.97%, FNR of 2.75%, FPR of 1.04%, FDR of 2.52%, FOR of 1.03%, MCC of 96.22%, BA and GM of 98.10% and 98.09%, respectively. Below Table 18 shows the performance evaluation of the InceptionV3 model regarding different performance evaluation parameters, including accuracy, MCR, precision, sensitivity, specificity, F1-score, and IoU of each class on different optimizers. In contrast, Table 19 shows the overall performance of the InceptionV3 model regarding accuracy, MCR, precision, sensitivity, specificity, F1-score, and IoU, and Table 20 shows the overall performance of the InceptionV3 model regarding NPV, FNR, FPR, FDR, FOR, MCC, BA, and (GM) on different optimizers. Table 18 Performance of InceptionV3 model regarding each class on different optimizers. Optimizers Image Class Accuracy MCR Precision Sensitivity Specificity F1-Score I0U Adam Glioma 99.35% 0.65% 99.07% 99.53% 99.19% 99.30% 98.61% Meningioma 97.83% 2.17% 98.00% 92.45% 99.44% 95.15% 90.74% Pituitary 98.48% 1.52% 95.86% 99.29% 98.13% 97.54% 95.21% SGD Glioma 83.70% 16.30% 75.65% 95.79% 73.17% 84.54% 73.21% Meningioma 81.30% 18.70% 76.32% 27.36% 97.46% 40.28% 25.22% Pituitary 88.48% 11.52% 78.81% 85.00% 90.00% 81.79% 69.19% RMSProp Glioma 99.35% 0.65% 99.53% 99.07% 99.59% 99.30% 98.60% Meningioma 98.26% 1.74% 97.12% 95.28% 99.15% 96.19% 92.66% Pituitary 98.91% 1.09% 97.20% 99.29% 98.75% 98.23% 96.53% Nadam Glioma 99.78% 0.22% 100.00% 99.53% 100.00% 99.77% 99.53% Meningioma 98.04% 1.96% 97.09% 94.34% 99.15% 95.69% 91.74% Pituitary 98.26% 1.74% 95.83% 98.57% 98.13% 97.18% 94.52% Table 18 shows the comparison of different evaluation parameters on different optimizers of the InceptionV3 model regarding each class. The model achieved the best results on RMSProp Optimizer, with an accuracy of 99.35%, 98.26%, and 98.91% for glioma, meningioma, and pituitary tumor classes, respectively, with MCRs of 0.65%, 1.74%, and 1.09% for each class, respectively. While the model achieved the lowest results on the SGD optimizer, achieving accuracy of 83.70%, 81.30%, and 88.48% for glioma, meningioma, and pituitary tumor classes, respectively, with MCRs of 16.30%, 18.70%, and 11.52% for each class respectively. Table 19 Overall performance of InceptionV3 model on different optimizers (Continue). Optimizers Accuracy MCR Precision Sensitivity Specificity F1-Score IoU Adam 98.55% 1.45% 97.64% 97.09% 98.92% 97.33% 94.85% SGD 84.49% 15.51% 76.92% 69.38% 86.88% 68.87% 55.87% RMSProp 98.84% 1.16% 97.95% 97.88% 99.17% 97.91% 95.93% Nadam 98.70% 1.30% 97.64% 97.48% 99.09% 97.55% 95.27% Table 19 shows the overall performance of the InceptionV3 model on different optimizers regarding accuracy, MCR, precision, sensitivity, specificity, F1-score, and IoU. On all optimizers, RMSProp Optimizer produced the best results, and the model achieved results such as accuracy of 98.84%, MCR of 1.16%, precision of 97.95%, sensitivity of 97.88%, specificity of 99.17%, F1-score of 97.91%, and IoU of 95.93%. While the SGD optimizer produced the worst results, and the model achieved an accuracy of 84.49% with an MCR of 15.51%, precision of 76.92%, sensitivity of 69.38%, specificity of 86.88%, F1-score of 68.87%, and IoU of 55.87%. While Adam and Nadam optimizers produced moderate results better than SGD and model achieved results on Adam with an accuracy of 98.55%, MCR of 1.45%, precision of 97.64%, sensitivity of 97.09%, specificity of 98.92%, F1-score of 97.33%, and IoU of 94.85%. while InceptionV3 achieved results on Nadam Optimizer as accuracy of 98.70%, MCR of 1.30%, precision of 97.64%, sensitivity of 97.48%, specificity of 99.09%, F1-score of 97.55%, and IoU of 95.27%. Table 20 Overall performance of the InceptionV3 model on different optimizers. Optimizers NPV FNR FPR FDR FOR MCC BA GM Adam 99.02% 2.91% 1.08% 2.36% 0.98% 96.07% 98.00% 97.98% SGD 90.07% 30.62% 13.12% 23.08% 9.93% 59.24% 78.13% 74.27% RMSProp 99.16% 2.12% 0.83% 2.05% 0.84% 97.04% 98.52% 98.52% Nadam 99.09% 2.52% 0.91% 2.36% 0.91% 96.58% 98.29% 98.28% Table 20 shows the overall performance of the InceptionV3 model regarding NPV, FNR, FDR, FOR, MCC, BA, and GM on different optimizers. As we see above, RMSProp Optimizer produced better results, Adam and Nadam produced moderate results, and SGD Optimizer produced the worst results. Regarding Table 18 performance parameters, the model achieved values on Adam Optimizer as NPV of 99.02%, FNR of 2.91%, FPR of 1.08%, FDR of 2.36%, FOR of 0.98%, MCC of 96.07%, BA of 98.00%, and GM of 97.98%. While on the SGD optimizer, the model achieved less results, as NPV of 90.07%, FNR of 30.62%, FPR of 13.12%, FDR of 23.08%, FOR of 9.93%, a MCC of 59.24%, a BA of 78.13%, and a GM of 74.27%. Further, the RMSProp model achieved a NPV of 99.16%, a FNR of 2.12%, FPR of 0.83%, FDR of 2.05%, FOR of 0.84%, MCC of 97.04%, BA and GM of 98.52%, and 98.52%, respectively. While on Nadam Optimizer, the model achieved results as NPV of 99.09%, FNR of 2.52%, FPR of 0.91%, FDR of 2.36%, FOR of 0.91%, MCC of 96.58%, BA and GM of 98.29% and 98.28%, respectively. Table 21 shows the performance evaluation of the MobileNetV1 model regarding different performance evaluation parameters, including accuracy, MCR, precision, sensitivity, specificity, F1-score, and IoU of each class on different optimizers, while Table 22 shows the overall performance of the MobileNetV1 model regarding accuracy, MCR, precision, sensitivity, specificity, F1-score, and IoU, and Table 23 shows the overall performance of the MobileNetV1 model regarding NPV, FNR, FPR, FDR, FOR, MCC, BA, and GM on different optimizers. Table 21 Performance of MobileNetV1 model regarding each class on different optimizers. Optimizers Image Class Accuracy MCR Precision Sensitivity Specificity F1-Score IoU Adam Glioma 99.35% 0.65% 99.07% 99.53% 99.19% 99.30% 98.61% Meningioma 98.04% 1.96% 98.99% 92.45% 99.72% 95.61% 91.59% Pituitary 98.70% 1.30% 95.89% 100.00% 98.13% 97.90% 95.89% SGD Glioma 91.30% 8.70% 91.43% 89.72% 92.68% 90.57% 82.76% Meningioma 87.17% 12.83% 74.74% 66.98% 93.22% 70.65% 54.62% Pituitary 93.70% 6.30% 85.81% 95.00% 93.13% 90.17% 82.10% RMSProp Glioma 99.57% 0.43% 100.00% 99.07% 100.00% 99.53% 99.07% Meningioma 98.48% 1.52% 98.06% 95.28% 99.44% 96.65% 93.52% Pituitary 98.91% 1.09% 96.55% 100.00% 98.44% 98.25% 96.55% Nadam Glioma 99.57% 0.43% 100.00% 99.07% 100.00% 99.53% 99.07% Meningioma 97.83% 2.17% 97.06% 93.40% 99.15% 95.19% 90.83% Pituitary 98.26% 1.74% 95.21% 99.29% 97.81% 97.20% 94.56% Table 21 shows the comparison of different evaluation parameters on different optimizers of the MobileNetV1 model regarding each class. The model achieved the best results on RMSProp Optimizer, with an accuracy of 99.57%, 98.48%, and 98.91% for glioma, meningioma, and pituitary tumor classes, respectively, with MCRs of 0.43%, 1.52%, and 1.09% for each class, respectively. While the model achieved the lowest results on the SGD optimizer, achieving an accuracy of 91.30%, 87.17%, and 93.70% for glioma, meningioma, and pituitary tumor classes, respectively, with MCRs of 8.70%, 12.83%, and 6.30% for each class respectively. Table 22 Overall performance of MobileNetV1 model on different optimizers (Continue). Optimizers Accuracy MCR Precision Sensitivity Specificity F1-Score IoU Adam 98.70% 1.30% 97.98% 97.33% 99.01% 97.60% 95.36% SGD 90.72% 9.28% 83.99% 83.90% 93.01% 83.79% 73.16% RMSProp 98.99% 1.01% 98.20% 98.12% 99.29% 98.14% 96.38% Nadam 98.55% 1.45% 97.42% 97.25% 98.99% 97.31% 94.82% Table 22 shows the overall performance of the MobileNetV1 model on different optimizers regarding accuracy, MCR, precision, sensitivity, specificity, F1-score, and IoU. On all optimizers, RMSProp Optimizer produced the best results, and the model achieved results such as accuracy of 98.99%, MCR of 1.01%, precision of 98.20%, sensitivity of 98.12%, specificity of 99.29%, F1-score of 98.14%, and IoU of 96.38%. While the SGD optimizer produced the worst results and the model achieved an accuracy of 90.72% with MCR of 9.28%, precision of 83.99%, sensitivity of 83.90%, specificity of 93.01%, F1-score of 83.79%, and IoU of 73.16%. While Adam and Nadam optimizers produced moderate results better than SGD and model achieved results on Adam as accuracy of 98.70%, MCR of 1.30%, precision of 97.98%, sensitivity of 97.33%, specificity of 99.01%, F1-score of 97.60%, and IoU of 95.36%. while MobileNetV1 achieved results on Nadam Optimizer as accuracy of 98.55%, MCR of 1.45%, precision of 97.42%, sensitivity of 97.25%, specificity of 98.99%, F1-score of 97.31%, and IoU of 94.82%. Table 23 Overall performance of MobileNetV1 model on different optimizers. Optimizers NPV FNR FPR FDR FOR MCC BA GM Adam 99.13% 2.67% 0.99% 2.02% 0.87% 96.42% 98.17% 98.14% SGD 93.11% 16.10% 6.99% 16.01% 6.89% 77.10% 88.45% 88.09% RMSProp 99.26% 1.88% 0.71% 1.80% 0.74% 97.40% 98.70% 98.69% Nadam 98.97% 2.75% 1.01% 2.58% 1.03% 96.24% 98.12% 98.10% Table 23 shows the overall performance of the MobileNetV1 model regarding NPV, FNR, FDR, FOR, MCC, BA, and GM on different optimizers. As we see above, RMSProp Optimizer produced better results, Adam and Nadam produced moderate results, and SGD Optimizer produced the worst results. Regarding Table 21 performance parameters, the model achieved values on Adam Optimizer as NPV of 99.13%, FNR of 2.67%, FPR of 0.99%, FDR of 2.02%, FOR of 0.87%, MCC of 96.42%, BA of 98.17%, and GM of 98.14%. While on the SGD optimizer, the model achieved less results, as NPV of 93.11%, FNR of 16.10%, FPR of 6.99%, FDR of 16.01%, FOR of 6.89%, a MCC of 77.10%, a BA of 88.45%, and a GM of 88.09%. Further, the RMSProp model achieved a NPV of 99.26%, a FNR of 1.88%, FPR of 0.71%, FDR of 1.80%, FOR of 0.74%, MCC of 97.40%, BA and GM of 98.70%, and 98.69%, respectively. While on Nadam Optimizer, the model achieved results, as NPV of 98.97%, FNR of 2.75%, FPR of 1.01%, FDR of 2.58%, FOR of 1.03%, MCC of 96.24%, BA and GM of 98.12% and 98.10%, respectively. Below Table 24 shows the performance evaluation of the DenseNet121 model regarding different performance evaluation parameters, including accuracy, MCR, precision, sensitivity, specificity, F1-score, and IoU of each class on different optimizers, while Table 25 shows the overall performance of the DenseNet121 model regarding accuracy, MCR, precision, sensitivity, specificity, F1-score, and IoU, and Table 26 shows the overall performance of the DenseNet121 model regarding NPV, FNR, FPR, FDR, FOR, MCC, BA, and GM on different optimizers. Table 24 Performance of DenseNet121 model regarding each class on different optimizers. Optimizers Image Class Accuracy MCR Precision Sensitivity Specificity F1-Score IoU Adam Glioma 99.35% 0.65% 100.00% 98.60% 100.00% 99.29% 98.60% Meningioma 98.70% 1.30% 98.08% 96.23% 99.44% 97.14% 94.44% Pituitary 98.91% 1.09% 96.55% 100.00% 98.44% 98.25% 96.55% SGD Glioma 89.13% 10.87% 91.84% 84.11% 93.50% 87.80% 78.26% Meningioma 84.57% 15.43% 69.23% 59.43% 92.09% 63.96% 47.01% Pituitary 90.65% 9.35% 78.03% 96.43% 88.13% 86.26% 75.84% RMSProp Glioma 99.13% 0.87% 99.53% 98.60% 99.59% 99.06% 98.14% Meningioma 97.83% 2.17% 96.15% 94.34% 98.87% 95.24% 90.91% Pituitary 98.70% 1.30% 96.53% 99.29% 98.44% 97.89% 95.86% Nadam Glioma 99.57% 0.43% 99.53% 99.53% 99.59% 99.53% 99.07% Meningioma 98.70% 1.30% 99.02% 95.28% 99.72% 97.12% 94.39% Pituitary 99.13% 0.87% 97.22% 100.00% 98.75% 98.59% 97.22% Table 24 shows the comparison of different evaluation parameters on different optimizers of the DenseNet121 model regarding each class. The model achieved the best results on Nadam Optimizer, with an accuracy of 99.57%, 98.70%, and 99.13% for glioma, meningioma, and pituitary tumor classes, respectively, with MCRs of 0.43%, 1.30%, and 0.87% for each class, respectively. While the model achieved the lowest results on the SGD optimizer, achieving accuracy of 89.13%, 84.57%, and 90.65% for glioma, meningioma, and pituitary tumor classes, respectively, with MCRs of 10.87%, 15.43%, and 9.35% for each class respectively. Table 25 Overall performance of DenseNet121 model on different optimizers (Continue). Optimizers Accuracy MCR Precision Sensitivity Specificity F1-Score IoU Adam 98.99% 1.01% 98.21% 98.27% 99.29% 98.23% 96.53% SGD 88.12% 11.88% 79.70% 79.99% 91.24% 79.34% 67.04% RMSProp 98.55% 1.45% 97.40% 97.41% 98.97% 97.40% 94.97% Nadam 99.13% 0.87% 98.59% 98.27% 99.35% 98.41% 96.89% Table 25 shows the overall performance of the DenseNet121 model on different optimizers regarding accuracy, MCR, precision, sensitivity, specificity, F1-score, and IoU. On all optimizers, Nadam Optimizer produced the best results, and the model achieved results such as accuracy of 99.13%, MCR of 0.87%, precision of 98.59%, sensitivity of 98.27%, specificity of 99.35%, F1-score of 98.41%, and IoU of 96.89%. While the SGD optimizer produced the worst results and the model achieved an accuracy of 88.12% with MCR of 11.88%, precision of 79.70%, sensitivity of 79.99%, specificity of 91.24%, F1-score of 79.34%, and IoU of 67.04%. While Adam and RMSProp optimizers produced moderate results better than SGD and model achieved results on Adam as accuracy of 98.99%, MCR of 1.01%, precision of 98.21%, sensitivity of 98.27%, specificity of 99.29%, F1-score of 98.23%, and IoU of 96.53%. while DenseNet121 achieved results on RMSProp Optimizer as accuracy of 98.55%, MCR of 1.45%, precision of 97.40%, sensitivity of 97.41%, specificity of 98.97%, F1-score of 97.40%, and IoU of 94.97%. Table 26 Overall performance of DenseNet121 model on different optimizers. Optimizers NPV FNR FPR FDR FOR MCC BA GM Adam 99.22% 1.73% 0.71% 1.79% 0.78% 97.54% 98.78% 98.78% SGD 91.24% 20.01% 8.76% 20.30% 8.76% 71.55% 85.61% 84.95% RMSProp 98.93% 2.59% 1.03% 2.60% 1.07% 96.36% 98.19% 98.18% Nadam 99.40% 1.73% 0.65% 1.41% 0.60% 97.66% 98.81% 98.80% Table 26 shows the overall performance of the DenseNet121 model regarding NPV, FNR, FDR, FOR, MCC, BA, and GM on different optimizers. As we see above, Nadam Optimizer produced better results, Adam and RMSProp produced moderate results, and SGD Optimizer produced the worst results. Regarding Table 24 performance parameters, the model achieved values on Adam Optimizer as NPV of 99.22%, FNR of 1.73%, FPR of 0.71%, FDR of 1.79%, FOR of 0.78%, MCC of 97.54%, BA of 98.78%, and GM of 98.78%. While on the SGD optimizer, the model achieved less results, as NPV of 91.24%, FNR of 20.01%, FPR of 8.76%, FDR of 20.30%, FOR of 8.76%, a MCC of 71.55%, a BA of 85.61%, and a GM of 84.95%. Further, the RMSProp model achieved a NPV of 98.93%, a FNR of 2.59%, FPR of 1.03%, FDR of 2.60%, FOR of 1.07%, MCC of 96.36%, BA and GM of 98.19%, and 98.18%, respectively. While on Nadam Optimizer, the model achieved results, as NPV of 99.40%, FNR of 1.73%, FPR of 0.65%, FDR of 1.14%, FOR of 0.60%, MCC of 97.66%, BA and GM of 98.81% and 98.80%, respectively. Additionally, Table 27 shows the performance evaluation of the EfficientNetB0 model regarding different performance evaluation parameters, including accuracy, MCR, precision, sensitivity, specificity, F1-score, and IoU of each class on different optimizers, while Table 28 shows the overall performance of the EfficientNetB0 model regarding accuracy, MCR, precision, sensitivity, specificity, F1-score, and IoU, and Table 29 shows the overall performance of the EfficientNetB0 model regarding NPV, FNR, FPR, FDR, FOR, MCC, BA, and GM on different optimizers. Table 27 Performance of EfficientNetB0 model regarding each class on different optimizers. Optimizers Image Class Accuracy MCR Precision Sensitivity Specificity F1-Score IoU Adam Glioma 100.00% 0.00% 100.00% 100.00% 100.00% 100.00% 100.00% Meningioma 98.70% 1.30% 100.00% 94.34% 100.00% 97.09% 94.34% Pituitary 98.70% 1.30% 95.89% 100.00% 98.13% 97.90% 95.89% SGD Glioma 58.48% 41.52% 52.96% 96.26% 25.61% 68.33% 51.89% Meningioma 76.52% 23.48% 37.50% 2.83% 98.59% 5.26% 2.70% Pituitary 73.26% 26.74% 63.49% 28.57% 92.81% 39.41% 24.54% RMSProp Glioma 99.35% 0.65% 99.53% 99.07% 99.59% 99.30% 98.60% Meningioma 98.26% 1.74% 100.00% 92.45% 100.00% 96.08% 92.45% Pituitary 98.04% 1.96% 93.96% 100.00% 97.19% 96.89% 93.96% Nadam Glioma 98.91% 1.09% 98.16% 99.53% 98.37% 98.84% 97.71% Meningioma 97.61% 2.39% 100.00% 89.62% 100.00% 94.53% 89.62% Pituitary 98.26% 1.74% 94.59% 100.00% 97.50% 97.22% 94.59% Table 27 shows the comparison of different evaluation parameters on different optimizers of the EfficientNetB0 model regarding each class. The model achieved the best results on Adam Optimizer, with an accuracy of 100%, 98.70%, and 98.70% for glioma, meningioma, and pituitary tumor classes, respectively, with MCRs of 0.0%, 1.30%, and 1.30% for each class, respectively. While the model achieved the lowest results on the SGD optimizer, achieving accuracy of 58.48%, 76.52%, and 73.26% for glioma, meningioma, and pituitary tumor classes, respectively, with MCRs of 41.52%, 23.48%, and 26.74% for each class respectively. Table 28 Overall performance of EfficientNetB0 model on different optimizers (Continue). Optimizers Accuracy MCR Precision Sensitivity Specificity F1-Score IoU Adam 99.13% 0.87% 98.63% 98.11% 99.38% 98.33% 96.74% SGD 69.42% 30.58% 51.32% 42.55% 72.34% 37.67% 26.38% RMSProp 98.55% 1.45% 97.83% 97.17% 98.93% 97.42% 95.01% Nadam 98.26% 1.74% 97.58% 96.39% 98.62% 96.86% 93.97% Table 28 shows the overall performance of the EfficientNetB0 model on different optimizers regarding accuracy, MCR, precision, sensitivity, specificity, F1-score, and IoU. On all optimizers, Adam Optimizer produced the best results, and the model achieved results such as accuracy of 99.13%, MCR of 0.87%, precision of 98.63%, sensitivity of 98.11%, specificity of 99.38%, F1-score of 98.33%, and IoU of 96.74%. While the SGD optimizer produced the worst results and the model achieved an accuracy of 69.42% with MCR of 30.58%, precision of 51.32%, sensitivity of 42.55%, specificity of 72.34%, F1-score of 37.67%, and IoU of 26.38%. While Nadam and RMSProp optimizers produced moderate results better than SGD and model achieved results on RMSProp as accuracy of 98.55%, MCR of 1.45%, precision of 97.83%, sensitivity of 97.17%, specificity of 98.93%, F1-score of 97.42%, and IoU of 95.01%. while EfficientNetB0 achieved results on Nadam Optimizer as accuracy of 98.26%, MCR of 1.74%, precision of 97.58%, sensitivity of 96.39%, specificity of 98.62%, F1-score of 98.86%, and IoU of 93.97%. Table 29 Overall performance of EfficientNetB0 model on different optimizers. Optimizers NPV FNR FPR FDR FOR MCC BA GM Adam 99.44% 1.89% 0.63% 1.37% 0.56% 97.54% 98.74% 98.73% SGD 80.25% 57.45% 27.66% 48.68% 19.75% 20.39% 57.45% 39.28% RMSProp 98.99% 2.83% 1.07% 2.17% 1.01% 96.16% 98.05% 98.02% Nadam 98.86% 3.61% 1.38% 2.42% 1.14% 95.17% 97.50% 97.45% Table 29 shows the overall performance of the EfficientNetB0 model regarding NPV, FNR, FDR, FOR, MCC, BA, and GM on different optimizers. As we see above, Adam Optimizer produced better results, Nadam and RMSProp produced moderate results, and SGD Optimizer produced the worst results. Regarding Table 27 performance parameters, the model achieved values on Adam Optimizer as NPV of 99.44%, FNR of 1.89%, FPR of 0.63%, FDR of 1.37%, FOR of 0.56%, MCC of 97.54%, BA of 98.74%, and GM of 98.73%. While on the SGD optimizer, the model achieved less results, as NPV of 80.25%, FNR of 57.45%, FPR of 27.66%, FDR of 48.68%, FOR of 19.75%, a MCC of 20.39%, a BA of 57.45%, and a GM of 39.28%. Further, the RMSProp model achieved a NPV of 98.99%, a FNR of 2.83%, FPR of 1.07%, FDR of 2.17%, FOR of 1.01%, MCC of 96.16%, BA and GM of 98.05%, and 98.02%, respectively. While on Nadam Optimizer, the model achieved results, as NPV of 98.86%, FNR of 3.61%, FPR of 1.38%, FDR of 2.42%, FOR of 1.14%, MCC of 95.17%, BA and GM of 97.50% and 97.45%, respectively. Further, Tables 30 and 31 show the overall performance of the models in terms of achieving the best results with respect to optimizers regarding different statistical performance evaluation parameters that were already discussed above. Table 30 Overall Performance of the Models Regarding Best Results Producing Optimizers (Continue). Model Architecture Optimizers Accuracy MCR Precision Sensitivity Specificity F1-Score IoU AlexNet Adam 95.94% 4.06% 94.38% 92.29% 96.55% 93.19% 87.47% VGG16 Nadam 98.41% 1.59% 97.01% 97.56% 98.89% 97.27% 94.72% VGG19 Adam 99.13% 0.87% 98.58% 98.43% 99.33% 98.50% 97.06% ResNet50 Adam 98.41% 1.59% 97.33% 97.02% 98.82% 97.14% 94.49% Xception RMSProp 98.99% 1.01% 98.20% 98.03% 99.29% 98.10% 96.30% Inception V3 RMSProp 98.84% 1.16% 97.95% 97.88% 99.17% 97.91% 95.93% DenseNet121 Nadam 99.13% 0.87% 98.59% 98.27% 99.35% 98.41% 96.89% MobileNetV1 RMSProp 98.99% 1.01% 98.20% 98.12% 99.29% 98.14% 96.38% EfficientNetB0 Adam 99.13% 0.87% 98.63% 98.11% 99.38% 98.33% 96.74% Table 31 Overall Performance of the Models Regarding Best Results Producing Optimizers. Model Architecture Optimizers NPV FNR FPR FDR FOR MCC BA GM AlexNet Adam 97.16% 7.71% 3.45% 5.62% 2.84% 89.20% 94.42% 94.29% VGG16 Nadam 98.74% 2.44% 1.11% 2.99% 1.26% 96.38% 98.23% 98.23% VGG19 Adam 99.35% 1.57% 0.67% 1.42% 0.65% 97.78% 98.88% 98.88% ResNet50 Adam 98.85% 2.98% 1.18% 2.67% 1.15% 95.87% 97.92% 97.90% Xception RMSProp 99.29% 1.97% 0.71% 1.80% 0.71% 97.33% 98.66% 98.65% Inception V3 RMSProp 99.16% 2.12% 0.83% 2.05% 0.84% 97.04% 98.52% 98.52% DenseNet121 Nadam 99.40% 1.73% 0.65% 1.41% 0.60% 97.66% 98.81% 98.80% MobileNetV1 RMSProp 99.26% 1.88% 0.71% 1.80% 0.74% 97.40% 98.70% 98.69% EfficientNetB0 Adam 99.44% 1.89% 0.63% 1.37% 0.56% 97.54% 98.74% 98.73% In the above Tables 30 and 31 , the overall results of different models regarding different optimizers are shown with respect to the different performance evaluation parameters shown. All the details of the overall performance of the models regarding different optimizers are already discussed in the above tables, while the AlexNet, VGG19, ResNet50, and EfficientNetB0 models produced the overall best results on Adam optimizer by achieving an overall accuracy of 95.94%, 99.13%, 98.41%, and 99.13% with MCRs of 4.06%, 0.87%, 1.59%, and 0.87%, respectively. While VGG16 and DenseNet121 produced the best results on Nadam Optimizer by achieving accuracy of 98.41% and 99.13% with MCRs of 1.59% and 0.87%, respectively, Further, Xception, InceptionV3, and MobileNetV1 produced better results on the RMSProp optimizer, achieving an overall accuracy of 98.99%, 98.84%, and 98.99% with MCRs of 1.01%, 1.16%, and 1.88%, respectively. None of the pre-trained models produced better results on the SGD optimizer, and among all the models regarding optimizers, the Adam optimizer works best, and four different models produced better results on it. While among all the different models, three produced almost the same results on different optimizers, there was a slight difference in their performance evaluation parameters. VGG19 on Adam Optimizer, DenseNet121 on Nadam Optimizer, and EfficientNetB0 on Adam Optimizer produce the same accuracy and MCR, achieving an accuracy of 99.13% and an MCR of 0.87%, respectively. While the EfficientNetB0 model produced better precision and specificity than other models at 98.63% and 99.38%, respectively, VGG19 produced better sensitivity, F1-score, and IoU than other models at 98.43%, 98.50%, and 97.06%, respectively. While EfficientNetB0 produced better results on NPV at 99.44%, which is higher than other models achieved values, and better results on FPR, FDR, and FOR as 0.63%, 1.37%, and 0.56%, respectively, which is less than other models achieved values. Further, VGG19 produced better results on FNR, MCC, BA, and GM, with a 1.57% FNR, which is less than other models achieved, and 97.78%, 98.88%, and 98.88%, respectively, which are higher than other models’ values. Overall, we can predict that the VGG19 and EfficientNetB0 models achieved better results on the Adam optimizer. Further, Fig. 3 shows the overall performance of different pre-trained models with the best result-producing optimizer regarding accuracy and MCR. Further, the confusion matrix (CM) of the best results-producing optimizers on deep learning models is shown in Fig. 4 . Figure 4 shows the confusion matrices of all the pre-trained models with the best results, producing optimizers. The confusion matrix provides detailed counts of correct and incorrect predictions across all classes, which give insights into the model overall performance. Figure 5 shows the receiver operating character (ROC) curves of the VGG19 and EfficientNetB0 models, which achieved the overall best performances among different pre-trained models on different optimizers, and this curve represents the trade-off between true and false positive rates across all possible threshold classifications by summarizing performance in the area under curve (AUC) metric. Figure 5 shows the results of the VGG19 and EfficientNetB0 models on Adam optimizers in the form of ROC curves, respectively, while the VGG19 model produced AUC for all three classes as 1.00, 0.99, and 1.00 for pituitary, meningioma, and glioma classes, respectively, while the EfficientNetB0 model produced AUC for all three classes as 1.00, which shows the superiority of these two models on Adam optimizers over other models. Further, below figure 6 is the testing result of the best-producing deep learning models with their testing accuracy, actual and predicted class labels against three brain tumor classes, respectively. So above Fig. 6 shows the testing results of glioma, meningioma, and pituitary tumor with a testing accuracy of 97.61% with actual and predicted labels that show the ability of deep learning-based approaches to diagnose diseases accurately in humans to cure their problems and save the human’s life. From above all the models, EfficientNetB0 and VGG19 with Adam Optimizer perform well just because of the combined benefits of using both momentum, which accelerates the SGD by focusing on directionally consisting updates, and RMSProp, which adjusts the learning rate based on recenet gradient. DenseNet121 with Nadam Optimizor performs well just due to the incorporation of Nesterov momentum, which looks ahead by computing the gradient at the predicted future position. Adam optimizer overcomes issues like slow convergence or vanishing gradient problems, which allow the VGG19 model to perform well in classification tasks as well as for EfficientNetB0, while Nadam’s predictive updates fit well with densely connected layers of DenseNet121 that benefit from more accurate gradient updates, which allow the model to reuse features and parameters that enhance the performance on classification. Further, the below Table 32 shows the comparison of the proposed best result achievable model with other state-of-the-art approaches grading different statistical performance evaluation parameters. Table 32 Proposed model result comparison with other state-of-the-art approaches. Ref Techniques Dataset Accuracy MCR Precision Sensitivity /Recall Specificity F1-Score [ 8 ] 23-CNN layers with VGG16 Figshare 97.8% 2.2% 96.5% 96.4% Nill 96.4% [ 9 ] PDCNN Figshare 97.60% 2.40% 97% 97% Nill 97% [ 11 ] Ensemble ViTs Figshare 98.70% 1.30% Nill 97.78% 99.42% Nill [ 12 ] DenseNet41-cornerNet Figshare 98.8% 1.20% Nill Nill Nill Nill [ 13 ] Multimodal-SVM Figshare 98.92% 1.08% Nill 98.82% 99.02% Nill [ 14 ] VGGNet Figshare 98.93% 1.07% 99.11% 98.68% 99.13% Nill [ 15 ] DSRUF + HOG Figshare 90.27% 9.73% Nill Nill Nill Nill [ 16 ] ResNet50 - Adadelta Figshare 99.02% 0.98% Nill Nill Nill Nill Khushi et al. VGG19 - Adam Figshare 99.13% 0.87% 98.58% 98.43% 99.33% 98.50% DenseNet121 - Nadam Figshare 99.13% 0.87% 98.59% 98.27% 99.35% 98.41% EfficientNetB0 - Adam Figshare 99.13% 0.87% 98.63% 98.11% 99.38% 98.33% Table 32 shows the comparison of our best achievable results models regarding optimizers with other state-of-the art approaches. From above all models on four different optimizers, VGG19 with Adam, DenseNet121 with Nadam, and Effi-cientNetB0 with Adam optimizer achieved the best results in terms of accuracy and miss classification rate that all other state-of-the-art approaches by achieving 99.13% accuracy with 0.87% MCR. VGG19-Adam achieved the best sensitivity or recall value of 98.43% and the F1-Score value of 98.50%, which is better than our other models as well as with other approaches, including [ 8 ], [ 9 ], [ 11 ], and all other Nill in terms of sensitivity and better than other approaches including [ 8 ], [ 9 ], and all other Nill in terms of F1-score, respectively. Further, EfficientNetB0-Adam achieved the best result in terms of precision and specificity by achieving values of 98.63% and 99.38%, respectively, which is better than [ 8 ], [ 9 ], and all other Nill value approaches in terms of precision and better than [ 13 ], [ 14 ], and all other Nill approaches in terms of specificity, respectively. Further, all three models achieved the best accuracy and low MCR with other state-of-the-art approaches mentioned in Table 32 , which shows the superiority of our proposed research work. The above three models produced the same results regarding accuracy and MCR, but the other performance evaluation parameters results are changed, and every model produced different results regarding each class, but unfortunately the average results of these three models are the same only on accuracy and MCR. The reason is that each model is highly robust and well optimized, and their ability to learn and generalize from the data is comparable across different architectures when applied to the same dataset, and different optimizers help models reach a similar level of optimization, which leads toward similar performance, particularly on well-defined datasets. Overall, each model performed differently against each class. Conclusions In recent decades, brain tumors have become one of the deadliest diseases, with survival rates decreasing over time. Diagnosing brain tumors has become increasingly challenging for paramedical staff. Early detection is crucial for improving survival rates. To address this, we utilized various pre-trained deep learning models with different optimizers to facilitate the early diagnosis of brain tumors. We evaluated the performance of deep learning models, including AlexNet, VGG16, VGG19, ResNet50, Xception, InceptionV3, DenseNet121, MobileNetV1, and EfficientNetB0, using optimizers such as Adam, SGD, RMSProp, and Nadam, on the publicly available Figshare brain tumor dataset. Among the models tested, VGG19 and EfficientNetB0 performed exceptionally well with the Adam optimizer, achieving an accuracy of 99.13% and a misclassification rate (MCR) of 0.87%. Both models also demonstrated strong performance in other statistical evaluation parameters and ROC curve analysis. EfficientNetB0 outperformed the others, achieving an area under the curve (AUC) of 100% for each tumor class and testing an accuracy of 97.61% on random images. Limitations and future work The main purpose of this research was to evaluate various pre-trained deep learning models with different optimizers for the accurate detection and diagnosis of brain tumors using the publicly available Figshare dataset. Additionally, we plan to explore more deep learning-based models on this and other brain tumor datasets, as well as for diagnosing other diseases in the human body, with the aim of improving human survival. Declarations Funding: This research work have no funding. Data Availability Statement: The datasets that have been used in this research work have been taken from an open-source library, Kaggle: https://www.kaggle.com/datasets/ashkhagan/figshare-brain-tumor-dataset & Simulation files used during the current study are available from the corresponding author upon reasonable request. Conflicts of Interest: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. Author Contribution: Conceptualization, Writing original draft, Experimentation: H.M.T.K; Supervision: T.M; Validation, Proof reading: I.N and T.M. References N. Varuna Shree and T. N. R. Kumar, “Identification and classification of brain tumor MRI images with feature extraction using DWT and probabilistic neural network,” Brain Informatics , vol. 5, no. 1, pp. 23–30, 2018, doi: 10.1007/s40708-017-0075-5. A. Bunevicius, K. Schregel, R. Sinkus, A. Golby, and S. Patz, “REVIEW: MR elastography of brain tumors,” NeuroImage Clin. , vol. 25, no. October 2019, 2020, doi: 10.1016/j.nicl.2019.102109. M. Santos and N. P. Rocha, “Medical Imaging Repository Contributions for Radiation Protection Key Performance Indicators,” Procedia Comput. Sci. , vol. 196, pp. 590–597, 2021, doi: 10.1016/j.procs.2021.12.053. I. Naseer, S. Akram, T. Masood, A. Jaffar, M. A. Khan, and A. Mosavi, “Performance Analysis of State-of-the-Art CNN Architectures for LUNA16,” Sensors , vol. 22, no. 12, 2022, doi: 10.3390/s22124426. H. M. T. Khushi, T. Masood, A. Jaffar, M. Rashid, and S. Akram, “Improved Multiclass Brain Tumor Detection via Customized Pretrained EfficientNetB7 Model,” IEEE Access , vol. 11, no. October, pp. 117210–117230, 2023, doi: 10.1109/ACCESS.2023.3325883. H. M. T. Khushi, T. Masood, A. Jaffar, S. Akram, and S. M. Bhatti, “Performance analysis of state-of-the-art CNN architectures for brain tumour detection,” Int. J. Imaging Syst. Technol. , vol. 34, no. 1, pp. 1–15, 2024, doi: 10.1002/ima.22949. M. A. Talukder et al. , “An efficient deep learning model to categorize brain tumor using reconstruction and fine-tuning,” Expert Syst. Appl. , vol. 230, no. August 2022, p. 120534, 2023, doi: 10.1016/j.eswa.2023.120534. M. S. I. Khan et al. , “Accurate brain tumor detection using deep convolutional neural network,” Comput. Struct. Biotechnol. J. , vol. 20, pp. 4733–4745, 2022, doi: 10.1016/j.csbj.2022.08.039. T. Rahman and M. S. Islam, “MRI brain tumor detection and classification using parallel deep convolutional neural networks,” Meas. Sensors , vol. 26, no. February, p. 100694, 2023, doi: 10.1016/j.measen.2023.100694. N. Ullah et al. , “An Effective Approach to Detect and Identify Brain Tumors Using Transfer Learning,” Appl. Sci. , vol. 12, no. 11, 2022, doi: 10.3390/app12115645. S. Tummala, S. Kadry, S. A. C. Bukhari, and H. T. Rauf, “Classification of Brain Tumor from Magnetic Resonance Imaging Using Vision Transformers Ensembling,” Curr. Oncol. 2022, Vol. 29, Pages 7498-7511 , vol. 29, no. 10, pp. 7498–7511, Oct. 2022, doi: 10.3390/CURRONCOL29100590. M. Nawaz et al. , “[94] Analysis of Brain MRI Images Using Improved 2021.pdf,” pp. 1–18, 2021. S. Maqsood, R. Damaševičius, and R. Maskeliūnas, “Multi-Modal Brain Tumor Detection Using Deep Neural Network and Multiclass SVM,” Med. , vol. 58, no. 8, 2022, doi: 10.3390/medicina58081090. P. P. Malla, S. Sahu, and A. I. Alutaibi, “Classification of Tumor in Brain MR Images Using Deep Convolutional Neural Network and Global Average Pooling,” Processes , vol. 11, no. 3, p. 679, 2023, doi: 10.3390/pr11030679. W. Ayadi, I. Charfi, W. Elhamzi, and M. Atri, “Brain tumor classification based on hybrid approach,” Vis. Comput. , vol. 38, no. 1, pp. 107–117, 2022, doi: 10.1007/s00371-020-02005-1. Ö. Polat and C. Güngen, “Classification of brain tumors from MR images using deep transfer learning,” J. Supercomput. , vol. 77, no. 7, pp. 7236–7252, 2021, doi: 10.1007/s11227-020-03572-9. M. Agarwal, G. Rani, A. Kumar, P. K. K, R. Manikandan, and A. H. Gandomi, “Deep learning for enhanced brain Tumor Detection and classification,” Results Eng. , vol. 22, no. December 2023, p. 102117, 2024, doi: 10.1016/j.rineng.2024.102117. A. Bhuvaneswari Ramakrishnan, M. Sridevi, S. K. Vasudevan, R. Manikandan, and A. H. Gandomi, “Optimizing brain tumor classification with hybrid CNN architecture: Balancing accuracy and efficiency through oneAPI optimization,” Informatics Med. Unlocked , vol. 44, p. 101436, Jan. 2024, doi: 10.1016/J.IMU.2023.101436. R. Sekaran, A. K. Munnangi, M. Ramachandran, and A. H. Gandomi, “3D brain slice classification and feature extraction using Deformable Hierarchical Heuristic Model,” Comput. Biol. Med. , vol. 149, p. 105990, Oct. 2022, doi: 10.1016/J.COMPBIOMED.2022.105990. A. Kumar, M. Ramachandran, A. H. Gandomi, R. Patan, S. Lukasik, and R. K. Soundarapandian, “A deep neural network based classifier for brain tumor diagnosis,” Appl. Soft Comput. , vol. 82, p. 105528, Sep. 2019, doi: 10.1016/J.ASOC.2019.105528. “Figshare Brain Tumor Dataset.” Accessed: Jun. 15, 2024. [Online]. Available: https://www.kaggle.com/datasets/ashkhagan/figshare-brain-tumor-dataset K. R. Avery et al. , “ImageNet classification with deep convolutional neural networks,” Commun. ACM , vol. 7, no. 3, pp. 84–90, Apr. 2017, doi: 10.1145/3065386. K. Simonyan and A. Zisserman, “Very deep convolutional networks for large-scale image recognition,” 3rd Int. Conf. Learn. Represent. ICLR 2015 - Conf. Track Proc. , pp. 1–14, 2015. K. He, X. Zhang, S. Ren, and J. Sun, “Deep Residual Learning for Image Recognition”, Accessed: Jun. 16, 2024. [Online]. Available: http://image-net.org/challenges/LSVRC/2015/ F. Chollet, “Xception: Deep Learning with Depthwise Separable Convolutions”. C. Szegedy, V. Vanhoucke, S. Ioffe, J. Shlens, and Z. Wojna, “Rethinking the Inception Architecture for Computer Vision,” Proc. IEEE Comput. Soc. Conf. Comput. Vis. Pattern Recognit. , vol. 2016-Decem, pp. 2818–2826, 2016, doi: 10.1109/CVPR.2016.308. A. G. Howard et al. , “MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications,” Apr. 2017, Accessed: Dec. 25, 2023. [Online]. Available: https://arxiv.org/abs/1704.04861v1 G. Huang, Z. Liu, L. Van Der Maaten, and K. Q. Weinberger, “Densely Connected Convolutional Networks”, Accessed: Jun. 16, 2024. [Online]. Available: https://github.com/liuzhuang13/DenseNet. M. Tan and Q. V. Le, “EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks,” Int. Conf. Mach. Learn. , 2019. K. Chandra, A. Xie, J. Ragan-Kelley, and E. MEIJER, “Gradient Descent: The Ultimate Optimizer,” Adv. Neural Inf. Process. Syst. , vol. 35, pp. 8214–8225, Dec. 2022. I. Bello, B. Zoph, V. Vasudevan, and Q. V. Le, “Neural Optimizer Search with Reinforcement Learning.” PMLR, pp. 459–468, Jul. 17, 2017. Accessed: Jun. 20, 2024. [Online]. Available: https://proceedings.mlr.press/v70/bello17a.html Z. Zhang, “Improved Adam Optimizer for Deep Neural Networks,” 2018 IEEE/ACM 26th Int. Symp. Qual. Serv. IWQoS 2018 , Jan. 2019, doi: 10.1109/IWQOS.2018.8624183. Q. Zhang et al. , “Boosting Adversarial Attacks with Nadam Optimizer,” Electron. 2023, Vol. 12, Page 1464 , vol. 12, no. 6, p. 1464, Mar. 2023, doi: 10.3390/ELECTRONICS12061464. H. M. Shahzad, S. M. Bhatti, A. Jaffar, M. Rashid, and S. Akram, “Multi-Modal CNN Features Fusion for Emotion Recognition: A Modified Xception Model,” IEEE Access , vol. 11, no. September, pp. 94281–94289, 2023, doi: 10.1109/ACCESS.2023.3310428. I. Naseer, S. Akram, T. Masood, A. Jaffar, M. A. Khan, and A. Mosavi, “Performance Analysis of State-of-the-Art CNN Architectures for LUNA16,” Sensors 2022, Vol. 22, Page 4426 , vol. 22, no. 12, p. 4426, Jun. 2022, doi: 10.3390/S22124426. R. Ali et al. , “Deep Learning for Sarcasm Identification in News Headlines,” Appl. Sci. , vol. 13, no. 9, 2023, doi: 10.3390/app13095586. I. Naseer, T. Masood, S. Akram, A. Jaffar, M. Rashid, and M. A. Iqbal, “Lung Cancer Detection Using Modified AlexNet Architecture and Support Vector Machine,” Comput. Mater. Contin. , vol. 74, no. 1, pp. 2039–2054, 2023, doi: 10.32604/cmc.2023.032927. W. Arshad et al. , “Cancer Unveiled: A Deep Dive Into Breast Tumor Detection Using Cutting-Edge Deep Learning Models,” IEEE Access , vol. 11, no. December, pp. 133804–133824, 2023, doi: 10.1109/ACCESS.2023.3335604. I. Naseer, S. Akram, T. Masood, M. Rashid, and A. Jaffar, “Lung Cancer Classification Using Modified U-Net Based Lobe Segmentation and Nodule Detection,” IEEE Access , vol. 11, no. June, pp. 60279–60291, 2023, doi: 10.1109/ACCESS.2023.3285821. D. Chicco and G. Jurman, “The advantages of the Matthews correlation coefficient (MCC) over F1 score and accuracy in binary classification evaluation,” BMC Genomics , vol. 21, no. 1, pp. 1–13, 2020, doi: 10.1186/s12864-019-6413-7. L. I. Kuncheva, Á. Arnaiz-González, J. F. Díez-Pastor, and I. A. D. Gunn, “Instance selection improves geometric mean accuracy: a study on imbalanced data classification,” Prog. Artif. Intell. , vol. 8, no. 2, pp. 215–228, 2019, doi: 10.1007/s13748-019-00172-4. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6937303","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":474859526,"identity":"9d3e557d-6c08-4487-a52e-c3d6bdb4275b","order_by":0,"name":"Hafiz Muhammad Tayyab Khushi","email":"data:image/png;base64,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","orcid":"","institution":"Superior University","correspondingAuthor":true,"prefix":"","firstName":"Hafiz","middleName":"Muhammad Tayyab","lastName":"Khushi","suffix":""},{"id":474859527,"identity":"a3126c62-ec1c-4901-9974-ad799e069dc8","order_by":1,"name":"Tehreem Masood","email":"","orcid":"","institution":"Superior University","correspondingAuthor":false,"prefix":"","firstName":"Tehreem","middleName":"","lastName":"Masood","suffix":""},{"id":474859528,"identity":"fb4186f6-34a0-4e16-bafb-ba117a14c0e7","order_by":2,"name":"Iftikhar Naseer","email":"","orcid":"","institution":"Superior University","correspondingAuthor":false,"prefix":"","firstName":"Iftikhar","middleName":"","lastName":"Naseer","suffix":""}],"badges":[],"createdAt":"2025-06-20 09:08:31","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6937303/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6937303/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":85370811,"identity":"2d428485-c53b-4397-864a-b1110b2f0465","added_by":"auto","created_at":"2025-06-25 07:29:11","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":322842,"visible":true,"origin":"","legend":"\u003cp\u003eMethodology workflow for brain tumor classification.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-6937303/v1/cf89247e16ae4ab7898d438b.png"},{"id":85371237,"identity":"3c8604d0-5d2e-474f-b3eb-4c79b2f27d8f","added_by":"auto","created_at":"2025-06-25 07:37:11","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":241604,"visible":true,"origin":"","legend":"\u003cp\u003eTyped of tumor in brain including glioma, meningioma, and pituitary tumors [17].\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-6937303/v1/63d986ce91a828d721849eab.png"},{"id":85370814,"identity":"efe15421-1783-42ce-a702-0bf73b0b996b","added_by":"auto","created_at":"2025-06-25 07:29:11","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":201885,"visible":true,"origin":"","legend":"\u003cp\u003eOverall best achievable results of each model w.r.t its optimizer.\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-6937303/v1/fa3d1689df26ffb560b31b98.png"},{"id":85370816,"identity":"412a6657-5e49-438e-ad87-3ff03540ed92","added_by":"auto","created_at":"2025-06-25 07:29:11","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":339472,"visible":true,"origin":"","legend":"\u003cp\u003eConfusion Martix of all pre-trained models with best results.\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-6937303/v1/aa5f84b48d667797363df13d.png"},{"id":85372224,"identity":"685400f5-8616-4f0f-95a4-4cacbee97620","added_by":"auto","created_at":"2025-06-25 07:45:11","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":289053,"visible":true,"origin":"","legend":"\u003cp\u003eROC curves for VGG19 and EfficientNetB0 model on Adam Optimizor.\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-6937303/v1/fa0d965bfe09d63d51df97a6.png"},{"id":85370815,"identity":"2f5fb778-7edc-42da-a8d9-13d0efe4770e","added_by":"auto","created_at":"2025-06-25 07:29:11","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":469692,"visible":true,"origin":"","legend":"\u003cp\u003eFigure 4: Testing result for glioma, meningioma, and pituitary tumor.\u003c/p\u003e","description":"","filename":"6.png","url":"https://assets-eu.researchsquare.com/files/rs-6937303/v1/185e74673ec355f21999940f.png"},{"id":85372552,"identity":"242d1e80-96aa-4fcb-aaa4-8ff6853898ff","added_by":"auto","created_at":"2025-06-25 07:53:16","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":5556277,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6937303/v1/cc3d1e42-89b1-4da8-a4b8-1aa7f9a1fa47.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"\u003cp\u003eOptimizer-Aware Deep Learning for Brain Tumor Classification: A Study Using AlexNet to EfficientNetB0\u003c/p\u003e","fulltext":[{"header":"Introduction","content":"\u003cp\u003eCancer is an abnormal, uncontrolled growth of cells and issues in the body. There are a lot of cancerous diseases in the human body, such as lung cancer, liver cancer, blood cancer, brain tumors, breast cancer, and many more, but as we know, the brain is the most important part of the body, and all the activities inside the body and the movements of the body parts are controlled by the brain. If there is some disturbance in the brain, all other body parts should definitely be in danger. Therefore, the leading disease inside a body is brain tumors around the globe, which can be essential to diagnose, detect, and localize at an early stage to save the patient's life. Brain tumors are broadly classified into two types: primary brain tumors are benign, and secondary brain tumors are malignant [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]. Most researchers believe benign is not a cancerous disease in the brain, which is some disturbance in the brain that can be cured by some environmental changes, while malignant is a cancerous part of the brain that is too dangerous for the life of humans because of its metastatic nature. Malignant cells can spread to any part of the body from the brain and vice versa [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. The World Health Organization (WHO) and the American Brain Tumor Association (ABTA) classify additional malignant tumors into three types: glioma tumors, which come from glial cells and nourish the brain's neurons; meningioma tumors, which come from the meninges and protect the brain and spinal cord's membranes; and pituitary tumors, which come from the pituitary gland at the base of the brain and produce hormones that the body uses for various functions, according to the WHO and the ABTA.\u003c/p\u003e \u003cp\u003eMoreover, to analyze the internal anatomy of the brain and to diagnose the tumor in the brain, different medical imaging modalities nowadays have been used, and a few of them are very popular, including X-rays, Computed tomography (CT-Scans), and Magnetic resonance imaging (MRI) based images [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e], X-rays provides the fractured bones details. In contrast, CT scans are used to provide cross-sectional images of the details of body parts from different angles. At the same time, MRI is more beneficial because it provides details regarding soft tissues and cells in the body, and cancer is all about the unusual growth of cells or tissues. In contrast, X-rays and CT scans produce some radiation during the examination of the body parts, which leads to some disturbance in the body with the passage of time. At the same time, MRI is more safer than other imaging modalities.\u003c/p\u003e \u003cp\u003eFurther, multiple computational approaches have been used to overcome the issue of brain tumor diagnosis, including machine learning, in which different algorithms are applied to an image dataset for the extraction of different features from the images, and based on those features or patterns, either a tumor exists, or a type of abnormality is predicted. However, due to the handcrafted nature of feature extraction and poor performance on large datasets, people move toward deep learning-based approaches, where CNN-based algorithms are applied that extract the best features from the dataset automatically [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]. The advantages of deep learning over machine learning-based algorithms include their superior capacity to handle large-scale data, better outcomes, and easier application, as well as their superior ability to extract hierarchical features and complicated pattern recognition. Deep learning-based methods are crucial in the medical field for detecting abnormalities in many bodily areas, particularly in the detection of brain tumors, breast cancer, and disorders related to lung cancer. It also helps to find any object detection problem, and it\u0026rsquo;s all about based on data sampling and their training. To overcome the issue of data sampling and resource consumption, pretrained models have their own importance because they have been trained on a larger dataset and leverage their learning feature ability and knowledge to analyze newly unseen small data samples for solving the current problem. In order to make the most of limited resources and cut down on training time, pre-trained or transfer learning based algorithms are commonly used [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. Further, the pretrained models have already been trained on a larger dataset and for a new unseen dataset; common features like color, edges, and textures have already been learned and transferred for the new dataset; and for extracting unique features from the unseen dataset, a few hyperparameters and layers have been added to the pre-trained models for extracting suitable information regarding the problem, which helps to overcome the resource problem; and time savings in terms of training the models from scratch. The main purpose is to diagnose the disturbance in the body or brain, and that transfer learning-based model produced better outputs regarding other approaches that helped radiologists make their decisions on time to secure the lives of patients.\u003c/p\u003e \u003cp\u003eThe key contributions of this research work are as follows:\u003c/p\u003e \u003cp\u003e \u003col\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eThe main thing that this study adds is that it uses multiple optimizers to validate pre-trained deep learning models and then compares how well they classify brain tumors.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eTo enhance resource management and feature extraction, a transfer learning-based model was employed to assess the performance of these models on the publicly available Figshare brain tumor dataset.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eA data augmentation approach was applied to capture more generic and specific features from a diverse range of images, addressing the issue of overfitting.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eSeveral metrics were used to assess the models' performance. These included precision, sensitivity, specificity, F1-score, accuracy, and intersection over union for misclassification rate.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003c/ol\u003e \u003c/p\u003e \u003cp\u003eHere is the structure of the research article: The current state-of-the-art approaches are reviewed in Section 2, the material and methodology are presented in Section 3, the results are shown in Section 4, conclusion, and limits and future work is exhibited in Sections 5 and 6, respectively.\u003c/p\u003e"},{"header":"Literature Review","content":"\u003cp\u003eBrain tumor diagnosis utilizing various machine learning and deep learning-based techniques has been the subject of numerous published research articles. Many of them consider the issue of detecting whether tumors exist or not. Many of them focus on the categorization of tumors into their types, and very few of them just focus on the localization of tumor regions in the brain. They have adopted different approaches to diagnose the disturbance in the brain, and a few of them are discussed below.\u003c/p\u003e \u003cp\u003eIn their study, Khushi et al. [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e] compared various deep learning algorithms with respect to optimizers for brain tumor detection in order to evaluate the model's efficacy using the open-source Br35H dataset. That was the goal of their five-algorithm proposal, which included ResNet50, AlexNet, VGG16, and Stochastic Gradient Descent (SGD), as well as Adam, root mean square propagation (RMSprop), and adaptive moment estimation (AM). In addition, several deep learning models, sensitivity, specificity, accuracy, and MCR (miss classification rate) are utilized as statistical performance evaluation metrics to assess the efficacy of models. Using the SGD optimizer, the AlexNet model outperforms the competition, reaching a total accuracy of 98.79% and a miss classification rate of 1.20%.\u003c/p\u003e \u003cp\u003eAnother research work was proposed by Khushi et al. [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e], they presented a research methodology for the efficient diagnosis of brain tumors using the proposed EfficientNetB7 model. They also tried different pre-trained models, and out of them, they found better results on the EfficientNetB7 model, so they decided to upgrade that model with the addition of some hyperparameters and layers. The Sartaj brain tumor dataset was obtained from Kaggle for this purpose. The dataset has four classes: glioma, meningioma, pituitary, and no tumor. Prior to the analysis, the images underwent preprocessing that included noise removal and cropping to remove unnecessary parts. In addition, they assessed the suggested model using a variety of performance metrics, such as accuracy and miss classification rate, and ultimately achieved a 98.97% accuracy rate and a 1.02% miss classification rate.\u003c/p\u003e \u003cp\u003eTalukder et al. [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e], presented a research work based on brain tumor categorization by different deep learning based pre-trained models using reconstruction and fine-tuning mechanisms. For this, they acquired a publicly available dataset from Figshare, which has three tumor classes and a total of 3064 images. To make the images more visible, they applied a sharp filter before going on to deep learning models. In order to implement the reconstruction principle, they shortened the layers following the activation layer and added an augmentation layer after the models' input layers. The four pre-trained models that were utilized were DenseNet201, Xception, ResNet50V2, and InceptionResNetV2. As compared to the other models, ResNet50V2 outperformed them all with a 99.68% accuracy, 99.49% precision, 99.78% recall, 99.64% F1-Score, and 0.32% MAE (mean absolute error).\u003c/p\u003e \u003cp\u003eAnother research project was proposed by Khan et al. [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e], also suggested a study to correctly identify brain cancers with the use of a deep learning model. For these purposes, they chose two different datasets: one is Figshare, which is a multiclass-based dataset with 3064 images, and the other is the Harvard Medical dataset, which contains 152 MRI images of two classes. They proposed a 23-layer-based CNN architecture, which caused an overfitting issue, and to overcome this, and they mirrored the planned 23-layer CNN architecture using a pre-trained VGG16 model. On the Figshare dataset, the suggested model attained an accuracy of 97.8% and a precision of 96.5%; on the Harvard medical dataset, it reached an accuracy of 100% with 100% precision, recall, and F1-Scor, respectively. In order to avoid overfitting in future work, it would be beneficial to test the suggested model alongside alternative deep learning models.\u003c/p\u003e \u003cp\u003eIn their study, Rahman et al. [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]. demonstrated how to use MRI images and a parallel deep convolutional neural network (PDCNN) model to detect brain cancers. They came up with a PDCNN model that could handle both local and global feature extraction. It did this by using two parallel CNN models on the images, which helped with overfitting. The model also included batch normalization and dropout regularizer layers. For these purposes, they chose three datasets, including the binary tumor classification dataset, the Figshare dataset, and the Sartaj Brain Tumor dataset. For the preprocessing phase, they resized the images with conversion to grayscale images and applied the data augmentation approach to make more data samples. Their proposed model achieved results of 97.33%, 97.60%, and 98.12% accuracy for the binary tumor dataset, the Figshare dataset, and the Sartaj brain tumor dataset, respectively.\u003c/p\u003e \u003cp\u003eUsing several pre-trained deep learning models for efficient tumor identification, Ullah et al. [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e] reported another study work focused on the categorization of brain tumors. For this, they have selected the publicly accessible Sartaj brain tumor dataset from Kaggle, comprising three tumor classes with one healthy image among four tumor types. Further, they adopted three tumor classes, and before passing to pre-trained models, they applied data augmentation to increase the samples of tumors. In addition, the images were scaled before being fed into deep learning models. The models then underwent modifications to their last three layers, which included a fully connected layer, softmax layers, and classification layers. Among the pre-trained models they utilized, InceptionResNetV2 outperformed the others, achieving an average accuracy of 98.91%, precision of 98.28%, recall of 99.75%, and F1-score of 99%. The other models included InceptionV3, Xception, ResNet18, ResNet50, ResNet101, ShuffleNet, DenseNet201, and MobileNetV2. Even with just two classes, it's important to test the proposed model on both a static and dynamic dataset in order to ensure accuracy.\u003c/p\u003e \u003cp\u003eTummala et al. [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e], presented research for the classification of tumors in the brain with the help of a vision transformer-ensembling approach. For this, they have chosen a well-reputed, publicly available Figshare dataset that has a total of 3064 images of three different tumor classes. For deep learning models, they chose the fine-tuned pre-trained ViT models with variants of B/16, B/32, L/16, and L/32. For result-oriented purposes, the ViT L/32 model achieved the best results, with an accuracy of 98.2% at 384 \u0026times; 384 image resolution. Further, for ensemble models of all ViT variants, the results achieved 98.7% accuracy at the same image resolution. Further, the ensemble model has to mention the validation accuracy of the models while also finding the other performance evaluation parameters on other brain tumor datasets.\u003c/p\u003e \u003cp\u003eA study by Nawaz et al. [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e] examined brain tumors and their accurate identification and classification using the DenseNet41-based CornerNet architecture. Figshare and the Brain MRI dataset were utilized for this objective. Their proposed model comprises three steps: first, they created the annotation to locate the exact region of interest, then further feature extraction was done by CornerNet with DenseNet41 as the backbone, and in the last one-stage detector, CornerNet was applied to locate and classify the tumor in the brain. On both datasets, their model achieved an average result of 98.8% and 98.5% accuracy for the Figshare and Brain MRI datasets, respectively. From a future perspective, the proposed model should be tested on a real-time dataset along with other disease predictions.\u003c/p\u003e \u003cp\u003eMaqsood et al. [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e] presented a study for brain tumor detection based on a multimodal deep neural network and a multiclass SVM approach; therefore, they chose two datasets from the open source library Kaggle: one is Brats2018 and the other is Figshare. After applying linear contrast stretching to extract edges, they used a 17-layer customized model for segmentation. For feature extraction, they used a modified MobileNetV2, and for brain tumor classification, they used multiclass SVM and M-SVM. On top of that, the model got 97.47% accuracy on the Brats2018 dataset and 98.92% accuracy on the Figshare dataset. The authors are compelled to utilize an existing model for segmentation and conduct classification using a deep learning model rather than a multiclass support vector machine, hence eliminating the need for seventeen layers.\u003c/p\u003e \u003cp\u003eAnother research work presented by Malla et al. [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e] for the classification of tumors in the brain with the help of MRI images, was done by applying CNN with a global average pooling layer. For these purposes, they have chosen the Figshare dataset from Kaggle and applied a pre-trained VGG16 model with a global average-pooling layer by freezing model layers to overcome the overfitting issue and banishing the gradient problem. In order to improve the signal-to-noise ratio, eliminate background noise, and maintain edge details, they boost picture contrast during the preprocessing phase. To improve the quality of the features extracted from photos by deep learning models, data augmentation is also used to improve the data samples. A sensitivity level of 98.68%, a specificity level of 99.13%, and a precision level of 99.11% were all obtained by the suggested model. If the author wants to know how well VGG16 performs on datasets, she should utilize other deep learning models; yet, the findings still could be better.\u003c/p\u003e \u003cp\u003eAyadi et al. [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e], presented a research work based on brain tumor classification using a hybrid approach. To achieve this, they analyzed data collected between 2005 and 2010 from two hospitals in China: Nan-Fang Hospital and General Hospital, which is named the Figshare dataset and has records of 233 patients against three tumor classes. The first step is image intensity normalization using min-max techniques. After further processing with DSURF and HOG, the features are input into the SVM classifier, which then sorts the tumors into three categories. A precision of 90.27 percent was attained by the suggested model. In addition, the outcome might be better, and a strategy based on deep learning would be ideal for automated feature extraction and categorization.\u003c/p\u003e \u003cp\u003ePolat et al. [\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e] published findings from their study on classifying brain tumors using deep learning models trained on magnetic resonance imaging (MR) scans. The Figshare dataset had three types of tumors including glioma, meningioma, and pituitary tumors, and was sourced from the open-source Kaggle library. Further, they passed the dataset to different pre-trained models with different optimizers by adding fully connected layers, and out of all the various deep learning models, the ResNet50 model with the Adadelta optimizer achieved the best result and got 99.02% accuracy. Further testing of deep learning models using various optimizers on the Figshare dataset is necessary to evaluate their performance on this multi-class tumor dataset for future perceptive.\u003c/p\u003e \u003cp\u003eAgarwal et al. [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e] presented research work for enhanced brain tumor classification and detection on a publicly available Figshare dataset. For this, they have adopted a two-phase approach, which consisted of enhancing the image contrast by applying ODTWCHE and the classification of the tumor by leaving the pre-trained InceptionV3 model for an accurate refine diagnostic process, and the proposed model achieved an accuracy of 98.89%. Ramakrishnan et al. [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e] proposed a hybrid CNN architecture-based model for brain tumor classification on a publicly available dataset to balance the accuracy and used VGG16, InceptionV3, DenseNet, and ResNet, and their efficiencies were improved by applying oneAPI optimization and achieved overall results with an accuracy of 96.2%. Proposing a deconvolutional residual network, Sekaran et al. [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e] proposed a deformable hierarchical heuristic-based model for 3D brain slice categorization RIDER dataset. The suggested model had an F1-score of 55% and an accuracy of 95% with 80% of precision. Kumar et al. [\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e] proposed a deep neural network first using weighted correlation-based feature selection technique and subsequently applied a multivariate neural network to reduce the miss classification rate as well as early brain tumor diagnosis improvements, so addressing the high false alarm rates. The below Table \u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e shows the literature review summary of different state-of-the art approaches.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eLiterature review survey on different computational approaches regarding brain tumor classification.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRef\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTechniques\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eDataset\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003ePerformance Evaluation\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eLimitations\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e[\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAlexNet with SGD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eBr35H\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAccuracy\u0026thinsp;=\u0026thinsp;98.79%,\u003c/p\u003e \u003cp\u003eMCR\u0026thinsp;=\u0026thinsp;1.20%\u003c/p\u003e \u003cp\u003eSensitivity\u0026thinsp;=\u0026thinsp;98.98%,\u003c/p\u003e \u003cp\u003eSpecificity\u0026thinsp;=\u0026thinsp;98.58%,\u003c/p\u003e \u003cp\u003eF1-Score\u0026thinsp;=\u0026thinsp;98.82%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eThe author should also have to test other deep learning models for the diagnosis of brain tumors.\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e[\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eModified EfficientNetB7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eSartaj\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAccuracy\u0026thinsp;=\u0026thinsp;98.97%,\u003c/p\u003e \u003cp\u003eMCR\u0026thinsp;=\u0026thinsp;1.02%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eThe proposed approach should be tested on different diseases and real datasets.\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e[\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eResNet50V2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eFigshare\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAccuracy\u0026thinsp;=\u0026thinsp;99.68%,\u003c/p\u003e \u003cp\u003eRecall\u0026thinsp;=\u0026thinsp;99.78%,\u003c/p\u003e \u003cp\u003eF1-Score\u0026thinsp;=\u0026thinsp;99.64%,\u003c/p\u003e \u003cp\u003ePrecision\u0026thinsp;=\u0026thinsp;99.49%,\u003c/p\u003e \u003cp\u003eMAE\u0026thinsp;=\u0026thinsp;0.32%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eThe proposed model should be used to evaluate different datasets and to diagnose other diseases.\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e[\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e23-CNN layers with VGG16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eFigshare\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAccuracy\u0026thinsp;=\u0026thinsp;97.8%,\u003c/p\u003e \u003cp\u003eRecall\u0026thinsp;=\u0026thinsp;96.4%,\u003c/p\u003e \u003cp\u003eF1-Score\u0026thinsp;=\u0026thinsp;96.4%,\u003c/p\u003e \u003cp\u003ePrecision\u0026thinsp;=\u0026thinsp;96.5%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eThe proposed model should be tried with other deep learning models to overcome the issue of overfitting.\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eHarvard Medical dataset\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAccuracy\u0026thinsp;=\u0026thinsp;100%,\u003c/p\u003e \u003cp\u003eRecall\u0026thinsp;=\u0026thinsp;100%,\u003c/p\u003e \u003cp\u003eF1-Score\u0026thinsp;=\u0026thinsp;100%,\u003c/p\u003e \u003cp\u003ePrecision\u0026thinsp;=\u0026thinsp;100%,\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003e[\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003ePDCNN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eBinary\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAccuracy\u0026thinsp;=\u0026thinsp;97.33%,\u003c/p\u003e \u003cp\u003eRecall\u0026thinsp;=\u0026thinsp;97.50%,\u003c/p\u003e \u003cp\u003eF1-Score\u0026thinsp;=\u0026thinsp;97.50%,\u003c/p\u003e \u003cp\u003ePrecision\u0026thinsp;=\u0026thinsp;97.50%,\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003eThere is still a need to overcome the issue of overfitting, and we should try any existing convolutional neural network model instead of making one from scratch.\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eFigshare\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAccuracy\u0026thinsp;=\u0026thinsp;97.60%,\u003c/p\u003e \u003cp\u003eRecall\u0026thinsp;=\u0026thinsp;97.00%,\u003c/p\u003e \u003cp\u003eF1-Score\u0026thinsp;=\u0026thinsp;97.00%,\u003c/p\u003e \u003cp\u003ePrecision\u0026thinsp;=\u0026thinsp;97.00%,\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eSartaj\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAccuracy\u0026thinsp;=\u0026thinsp;98.12%,\u003c/p\u003e \u003cp\u003eRecall\u0026thinsp;=\u0026thinsp;97.75%,\u003c/p\u003e \u003cp\u003eF1-Score\u0026thinsp;=\u0026thinsp;98.00%,\u003c/p\u003e \u003cp\u003ePrecision\u0026thinsp;=\u0026thinsp;98.00%,\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e[\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eInceptionResNetV2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eSartaj (three tumor classes)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAccuracy\u0026thinsp;=\u0026thinsp;98.91%,\u003c/p\u003e \u003cp\u003eRecall\u0026thinsp;=\u0026thinsp;99.75%,\u003c/p\u003e \u003cp\u003eF1-Score\u0026thinsp;=\u0026thinsp;99%,\u003c/p\u003e \u003cp\u003ePrecision\u0026thinsp;=\u0026thinsp;98.28%,\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eThere is still a need for improvements; instead of using three classes, there is still a need to validate the whole dataset on the proposed model, as well as on a real-time dataset.\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e[\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eEnsemble of ViTs\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eFigshare\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAccuracy\u0026thinsp;=\u0026thinsp;98.70%,\u003c/p\u003e \u003cp\u003eSensitivity\u0026thinsp;=\u0026thinsp;97.78%,\u003c/p\u003e \u003cp\u003eSpecificity\u0026thinsp;=\u0026thinsp;99.42%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eFurther, the ensemble model has to mention the validation accuracy of the models while also finding the other performance evaluation parameters on other brain tumor datasets.\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e[\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eDenseNet41-based CornerNet Framework\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eFigshare\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAccuracy\u0026thinsp;=\u0026thinsp;98.8%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eThere should be a need to test the proposed model on a real-time dataset along with other disease predictions\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eBrain MRI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAccuracy\u0026thinsp;=\u0026thinsp;98.5%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e[\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eMultimodal-M-SVM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eBrast2018\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAccuracy\u0026thinsp;=\u0026thinsp;97.47%,\u003c/p\u003e \u003cp\u003eSpecificity\u0026thinsp;=\u0026thinsp;97.04%,\u003c/p\u003e \u003cp\u003eDice Coef\u0026thinsp;=\u0026thinsp;96.71%,\u003c/p\u003e \u003cp\u003eSensitivity\u0026thinsp;=\u0026thinsp;97.22%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eInstead of using 17 layers, authors have to use an available segmentation model and perform classification using a deep learning model instead of a multiclass SVM.\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eFigshare\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAccuracy\u0026thinsp;=\u0026thinsp;98.92%,\u003c/p\u003e \u003cp\u003eSpecificity\u0026thinsp;=\u0026thinsp;99.02%,\u003c/p\u003e \u003cp\u003eDice Coef\u0026thinsp;=\u0026thinsp;99.87%,\u003c/p\u003e \u003cp\u003eSensitivity\u0026thinsp;=\u0026thinsp;98.82%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e[\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eVGGNet\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eFigshare\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAccuracy\u0026thinsp;=\u0026thinsp;98.93%,\u003c/p\u003e \u003cp\u003eSensitivity\u0026thinsp;=\u0026thinsp;98.68%,\u003c/p\u003e \u003cp\u003ePrecision\u0026thinsp;=\u0026thinsp;99.11%,\u003c/p\u003e \u003cp\u003eSpecificity\u0026thinsp;=\u0026thinsp;99.13%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eThe author should use different deep learning models instead of VGG16 to figure out the performance of that model on datasets, and still there is a need for improvements in the results.\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e[\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDSRUF\u0026thinsp;+\u0026thinsp;HOG\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eFigshare\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAccuracy\u0026thinsp;=\u0026thinsp;90.27%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eThere is still a need for improvements in the result, and there should be a need to use a deep learning-based approach for automatic feature extraction and classification.\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e[\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eResNet50 \u0026ndash;Adadelta\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eFigshare\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAccuracy\u0026thinsp;=\u0026thinsp;99.02%,\u003c/p\u003e \u003cp\u003eAUC\u0026thinsp;=\u0026thinsp;0.99%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eThere is still a need to check more deep learning models on the Figshare dataset with different optimizers to validate the performance of these multi-class tumor datasets, and more performance evaluation parameters are needed.\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eIn the literature review section, we have reviewed multiple research articles regarding brain tumor detection that have been published. Most research articles utilize the publicly accessible Figshare dataset, which comprises three tumor classifications including glioma, meningioma, and pituitary tumor has a record of 233 patients. Many researchers applied smaller, self-created models instead of using a well-defined existing model, and some of them used complex architectures for the detection of tumors in the brain, which further needed some improvements regarding models and results.\u003c/p\u003e"},{"header":"Materials and Methods","content":"\u003cp\u003eBrain tumors become the most lethal disease and the main reason for death in the world because of cancer, and still, there is a need for an accurate diagnosis or detection system that can predict tumors in their early stages to reduce the death ratio. In this research work, we have used different pre-trained models, including AlexNet, VGG16, VGG19, ResNet50, InceptionV3, Xception DenseNet121, MobileNetV1, and EfficientNetB0, with four different optimizers, utilizing Adam, Nadam, RMSprop, and SGD for brain tumor classification, as seen in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e3.1 - Data Acquisition\u003c/h2\u003e \u003cp\u003eThe overall procedure and workflow of the proposed model are described in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e. Further, before applying any computational approach, the first step is to obtain a dataset, which was taken from the open source library Kaggle and named Figshare brain tumor dataset [\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e], which has records of 233 patients and was taken from Nan-Fang Hospital and General Hospital, China, from 2005 to 2010. The dataset further has a total of 3064 images categorized into three tumor classes, including glioma, meningioma, and pituitary tumors. Out of 3064 images, the glioma tumor class contains 89 patients with 1426 images, while the meningioma tumor class contains 82 patients with 708 images, and the pituitary tumor class contains 62 patients with 930 images. All images are on three different planes, including axial, coronal, and sagittal, as shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003e3.2 - Data Preprocessing\u003c/h2\u003e \u003cp\u003eTo begin, the images were resized to 224 \u0026times; 224. In order to address the overfitting issues and improve the learning and extraction of features by deep learning models, a data augmentation approach was implemented prior to the dataset being entered into the training and validation phase. This approach generated an increased number of data samples from various perspectives. For data augmentation, a 7% rotation range was applied for rotating the images up to 7 degrees; further width and height shifts were applied to 5% for rotating the images, up to 5% of the total images horizontally and vertically. After that, a zoom range of 10% is applied to zoom in and out of the images by a maximum of 10%, and lastly, randomly flip the images horizontally and vertically. After that, dataset splitting was applied, and the dataset was split with an 85% ratio of training and a 15% ratio of validation, which is shown in Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e. Next, before passing data samples to different pre-trained models, label encoding is applied, which is essential for the model to process the labels correctly and to compute the loss during training and evaluation.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eData distribution of Figshare regarding training and validation w.r.t each class.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTumor Class\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTotal Samples\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eTraining Samples\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eValidation Samples\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eGlioma\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1426\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1212\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e214\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eMeningioma\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e708\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e602\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e106\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003ePituitary\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e930\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e790\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e140\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eTotal\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003e3064\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003e2604\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e\u003cb\u003e460\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003e3.3 - Pre-trained Deep learning model\u003c/h2\u003e \u003cp\u003eAdditionally, transfer learning-based models are implemented for the purpose of training and validation. As output layers, these models are equipped with the SoftMax classifier for the classification of brain tumors. They consist of a global average pooling layer, a dense layer with ReLu activation function, and a fully connected layer. Pre-trained models are preferred due to their capacity to extract valuable features from images and their prior training on a vast dataset. In contrast to a model that is formed from the ground up, the process of fine-tuning them on a new dataset typically necessitates less training time. When labeled data is scarce or inadequate, transfer learning enables the utilization of knowledge from a related domain that has data. This also helps when we have limited resources, alleviates the issue of overfitting, and enhances the model's capacity to generalize the unobserved data.\u003c/p\u003e \u003cp\u003eThe first model that we have selected is the AlexNet model, which was introduced by Alex Krizhevsky et al. in 2012 [\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e]. There were eight layers in total: five convolutional layers and three fully linked layers. The model also took advantage of the ReLU activation function, which overcomes the vanishing gradient issue, therefore enabling faster network training. To overcome the overfitting problem and cut the spatial dimensions, they also add the dropout and max-pooling layers. The ReLU activation function and dropout layer were added for the first time after the model was trained on 1000 image classes. VGG16 and VGG19 models were also used, constructed in 2014 by the Visual Geometry Group at the University of Oxford. Although the number of convolutional layers varies, the VGG16 and VGG19 models have a similar construction. VGG19 has 19 layers in total, including 16 convolutional layers, whereas VGG16 has 16 levels, 13 of which are convolutional. Both models use stochastic gradient descent (SGD) optimization for training and include three completely connected layers [\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eThe next model was ResNet50 [\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e], which was developed by Kaiming He et al. in 2015, and the key contribution of this model is residual learning to solve the vanishing gradient in deep network models. Using residual blocks, this model comprises of two or three convolutional layers with shortcut connections skipping one or more levels. There are a total 50 layers in the ResNet50 model, which is composed of 49 convolutional layers and an output layer that is divided into 16 residual blocks. The Further Xception model [\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e], which stands for Extreme Inception, was introduced by Francois Chollet and is based on depthwise separable convolutions. In the Xception model, residual connections are also used, which improves the gradient flow and training accuracy. This model replaces standard convolutions in Inception with depthwise separable convolution followed by pointwise convolution. InceptionV3 [\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e], also known as GoogleNet, was introduced by Christian Szegedy et al., which is the third version of the Inception family. This model allows or combines multiple convolutional operations and pooling operations in parallel, allowing the model to capture features at different scales. This model also employs batch normalization to speed up training and uses the RMSProp optimizer.\u003c/p\u003e \u003cp\u003eFurther, the MobileNetV1[\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e] model was selected, which was developed by Andrew G. Howard et al. The primary objective of this model's development was to reduce computational complexity and memory consumption while ensuring high performance. This model used the concepts of depthwise and pointwise separable convolution. There are a total of 13 depthwise separable convolutional layers, each followed by batch normalization, the ReLu6 activation function, and the RMSProp optimizer adopted for the training. Moreover, the DenseNet121 [\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e] model was introduced by Gao Huang et al, which belongs to the DenseNet family. DenseNet is comprised of dense blocks, in which each layer is connected to every other layer in a feed-forward manner. This results in a more efficient network and the utilization of features. The quantity of input feature maps placed between the dense blocks is reduced by the use of additional bottleneck layers and transition layers, which downsample the features, and the model is trained on SGD with momentum optimizers. The EfficientNetB0 [\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e] model is the baseline model of the EfficientNet family, introduced by Mingxing Tan and Quoc V. Le. The model is built utilizing a compound scaling technique whereby all dimensions of depth, width, and resolution are evenly scaled under control by a set of predetermined scaling factors. The model used mobile inverted bottleneck convolution blocks, which are composites of depthwise separable convolution, squeeze, excitation, and skip connections.\u003c/p\u003e \u003cp\u003eTherefore, for assessing the behavior and performance of the deep learning models with the addition of a global average pooling layer, which is responsible for lowering the dimensions of feature maps to a fixed size by taking the average value of each feature map to a fixed size across its spatial dimensions, the above mentioned pre-trained models have been trained on different optimizers, including Adam, SGD, RMSProp, and Nadam. The equations below illustrate how this mechanism operates.\u003cdiv id=\"Equ1\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ1\" name=\"EquationSource\"\u003e\n$$\\:GAP{\\left({F}_{in}\\right)}_{c}=\\:\\frac{1}{H\\:\\times\\:\\:W}\\:{\\sum\\:}_{i=1}^{H}{\\sum\\:}_{J=1}^{W}{F}_{in}(i,\\:j,\\:c)$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e1\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003eWhere \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{F}_{in}\\)\u003c/span\u003e\u003c/span\u003e is the feature map that comes as input from the base model with H \u0026times; W \u0026times; C, width, height, and channel on which the global average pooling layer is applied for reducing feature dimensions. Further, from the global average pooling layer, an output x is fed into a dense layer with 256 neurons and ReLu activation function to introduce learnable parameters, including biases and weights, which is responsible for enabling models to learn complex features or patterns and relationship from the aggregated features produced by global average pooling layer by also introducing non-linearity by ReLu activation function to learn complex, non-linear mapping from inputs to outputs.\u003cdiv id=\"Equ2\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ2\" name=\"EquationSource\"\u003e\n$$\\:x=GAP{\\left({F}_{in}\\right)}_{C}$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e2\u003c/div\u003e\u003c/div\u003e\u003cdiv id=\"Equ3\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ3\" name=\"EquationSource\"\u003e\n$$\\:h=ReLu({W}_{1\\:}\\bullet\\:\\:x+\\:{b}_{1})$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e3\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003eWhere x is the output vector from the global average pooling layer fed into a dense layer with 256 neurons and ReLu activation. \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:W\\)\u003c/span\u003e\u003c/span\u003e is the weight of a matrix size of 256, and \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{b}_{1}\\)\u003c/span\u003e\u003c/span\u003e is the bias vector of size 256, where h is the output from the dense layer which further fed into output layer with SoftMax activation function for the classification.\u003cdiv id=\"Equ4\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ4\" name=\"EquationSource\"\u003e\n$$\\:{y}_{j}=\\:\\frac{\\text{exp}\\left({({W}_{2\\:\\:}\\bullet\\:\\:h\\:+\\:{b}_{2})}_{j}\\right)}{{\\sum\\:}_{k=1}^{K}\\text{e}\\text{x}\\text{p}\\left({({W}_{2\\:\\:}\\bullet\\:\\:h\\:+\\:{b}_{2})}_{k}\\right)}\\:$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e4\u003c/div\u003e\u003c/div\u003e\u003cdiv id=\"Equ5\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ5\" name=\"EquationSource\"\u003e\n$$\\:y=SoftMax({W}_{2\\:}\\bullet\\:\\:ReLu\\left({W}_{1\\:}\\bullet\\:\\:GAP\\left({F}_{in}\\right)+\\:{b}_{1}\\right)+\\:{b}_{2})$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e5\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003eWhere \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{W}_{2\\:}\\)\u003c/span\u003e\u003c/span\u003eis the weight matrix of the output layer with size K \u0026times; 256, while \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{b}_{2}\\)\u003c/span\u003e\u003c/span\u003e is the bias vector of output layer with size k. y represents the output of the final layer after applying SoftMax activation function, which shows the probability for each class with size k.\u003c/p\u003e \u003cp\u003eAll the above-mentioned pre-trained models have been trained on different optimizers, including Adam, SGD, RMSProp, and Nadam, Deep learning or convolutional neural network models depend critically on optimizers since they update the neural network weights and reduce the loss function, therefore enabling the model to learn from the data and hence increase its performance over time. These optimizers also identify the set of parameters including weights and bias that reduce the loss function and track the variations between projected and real output. SGD is the first optimizer we have selected. [\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e], which updates the model parameters by taking small steps in the direction of the negative gradient of loss functions. It updates the weights for each training example individually, which can lead to noisy updates but often helps to escape local minima.\u003cdiv id=\"Equ6\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ6\" name=\"EquationSource\"\u003e\n$$\\:{\\theta\\:}_{t+1}\\:=\\:{\\theta\\:}_{t}-\\:\\eta\\:{\\nabla\\:}_{\\theta\\:}J\\left({\\theta\\:}_{t}\\right)$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e6\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003eWhere: \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\theta\\:}_{t}\\)\u003c/span\u003e\u003c/span\u003e are the parameters at time step \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:t\\)\u003c/span\u003e\u003c/span\u003e, while \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\eta\\:\\)\u003c/span\u003e\u003c/span\u003e is the learning rate \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\nabla\\:}_{\\theta\\:}J\\left({\\theta\\:}_{t}\\right)\\)\u003c/span\u003e\u003c/span\u003e is the gradient of the loss function \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:J\\)\u003c/span\u003e\u003c/span\u003e with respect to the parameters \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\theta\\:\\)\u003c/span\u003e\u003c/span\u003e. Subtracting the gradient's multiplication by the learning rate helps to update the parameters and guide them toward a lower loss.\u003c/p\u003e \u003cp\u003eThe next optimizer that we have chosen is RMSProp [\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e], which is an adaptive learning rate method designed for dealing with the issue of vanishing gradients and exploding gradients. It independently modulates the learning rate for every parameter. It first calculates the moving average of squared gradients then adjusts the settings.\u003cdiv id=\"Equ7\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ7\" name=\"EquationSource\"\u003e\n$$\\:{\\upsilon\\:}_{t}=\\:\\beta\\:{\\upsilon\\:}_{t-1}+(1-\\:\\beta\\:){\\left({\\nabla\\:}_{\\theta\\:}J\\right({\\theta\\:}_{t}\\left)\\:\\right)}^{2}$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e7\u003c/div\u003e\u003c/div\u003e\u003cdiv id=\"Equ8\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ8\" name=\"EquationSource\"\u003e\n$$\\:{\\theta\\:}_{t+1}\\:=\\:{\\theta\\:}_{t}-\\:\\frac{\\eta\\:}{\\sqrt{{\\upsilon\\:}_{t}+\\:ϵ}}\\:{\\nabla\\:}_{\\theta\\:}J\\left({\\theta\\:}_{t}\\right)\\:$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e8\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003eWhere: \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\upsilon\\:}_{t}\\)\u003c/span\u003e\u003c/span\u003e is the moving average of the squared gradients, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\beta\\:\\)\u003c/span\u003e\u003c/span\u003e is the decay rate, which is typically around 0.9, and \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:ϵ\\)\u003c/span\u003e\u003c/span\u003e is a small constant that is used to prevent division by zero. RMSProp dynamically changes the step size by inversely proportionalizing the learning rate for every parameter to the square root of the total of the recent squared gradient.\u003c/p\u003e \u003cp\u003eThe next optimizer that we have chosen is Adam [\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e], uses momentum in addition to RMSProp. It maintains gradients' uncentered variance (second moment) as well as their mean (first moment).\u003cdiv id=\"Equ9\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ9\" name=\"EquationSource\"\u003e\n$$\\:{m}_{t}=\\:{\\beta\\:}_{1}{m}_{t-1}+(1-\\:{\\beta\\:}_{1}){\\nabla\\:}_{\\theta\\:}J\\left({\\theta\\:}_{t}\\right)\\:$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e9\u003c/div\u003e\u003c/div\u003e\u003cdiv id=\"Equ10\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ10\" name=\"EquationSource\"\u003e\n$$\\:{\\upsilon\\:}_{t}=\\:{\\beta\\:}_{2}{\\upsilon\\:}_{t-1}+(1-\\:{\\beta\\:}_{2}){\\left({\\nabla\\:}_{\\theta\\:}J\\right({\\theta\\:}_{t}\\left)\\right)}^{2}\\:$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e10\u003c/div\u003e\u003c/div\u003e\u003cdiv id=\"Equ11\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ11\" name=\"EquationSource\"\u003e\n$$\\:{\\mathcal{m}}_{t}=\\:\\frac{{m}_{t}}{1-\\:{\\beta\\:}_{1}^{t}}$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e11\u003c/div\u003e\u003c/div\u003e\u003cdiv id=\"Equ12\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ12\" name=\"EquationSource\"\u003e\n$$\\:{\\mathcal{v}}_{t}=\\:\\frac{{\\upsilon\\:}_{t}}{1-\\:{\\beta\\:}_{2}^{t}}$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e12\u003c/div\u003e\u003c/div\u003e\u003cdiv id=\"Equ13\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ13\" name=\"EquationSource\"\u003e\n$$\\:{\\theta\\:}_{t+1}=\\:{\\theta\\:}_{t}-\\:\\frac{\\eta\\:}{\\sqrt{{\\mathcal{v}}_{t}}+\\:ϵ\\:}{\\mathcal{m}}_{t}$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e13\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003eWhere: \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{m}_{t}\\)\u003c/span\u003e\u003c/span\u003e and \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\upsilon\\:}_{t}\\)\u003c/span\u003e\u003c/span\u003e are the first and second-moment estimations, while \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\beta\\:}_{1}\\)\u003c/span\u003e\u003c/span\u003e and \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\beta\\:}_{2}\\)\u003c/span\u003e\u003c/span\u003e are the decay rates for the moment estimates commonly set to 0.9 and 0.999. Further, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\mathcal{m}}_{t}\\)\u003c/span\u003e\u003c/span\u003e and \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\mathcal{v}}_{t}\\)\u003c/span\u003e\u003c/span\u003e are the bias-corrected moment estimates, and \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:ϵ\\)\u003c/span\u003e\u003c/span\u003e is a small constant to prevent division by zero. Adam considers the mean and variance of the gradients to adjust the learning rates for every parameter, therefore enabling handling of both noisy and sparse gradients.\u003c/p\u003e \u003cp\u003eThe last optimizer that we have chosen is Nadam (Nesterov-accelerated adaptive moment estimation) [\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e], which is an extension of Adam that incorporates Nesterov momentum, which provides a more estimate of the gradient by looking ahead in the direction of momentum. Like Adam Optimizer, Nadam updates the biased first and second-moment estimations, computes the bias-corrected first and second-moment estimations, and then updates the parameter by incorporating the Nesterov momentum.\u003cdiv id=\"Equ14\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ14\" name=\"EquationSource\"\u003e\n$$\\:{\\theta\\:}_{t+1}=\\:{\\theta\\:}_{t}-\\:\\frac{\\eta\\:}{\\sqrt{{\\mathcal{v}}_{t}}+\\:ϵ\\:}({\\beta\\:}_{1}{\\mathcal{m}}_{t}+\\:\\frac{(1-\\:{\\beta\\:}_{1}){\\nabla\\:}_{\\theta\\:}J\\left({\\theta\\:}_{t}\\right)}{1-\\:{\\beta\\:}_{1}^{t}})\\:$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e14\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003eNadam improves upon Adam by incorporating the Nesterov accelerated gradient, which anticipates the direction of the gradient, leading to faster convergence and better performance. In the above-discussed optimizers, SGD is simple and works well for many problems but can be slow and stuck in local minima sometimes, while RMSProp adjusts the learning rate dynamically and is more suitable for non-stationary objectives. Further, Adam combined the advantages of both momentum and adaptive learning rates, which make him more robust and efficient for wide-ranging problems, and Nadam enhanced his performance by incorporating Nesterov momentum, which often provides better convergence and performance. Further, we have evaluated the different optimizers on different pre-trained models. Every optimizer has its own pros and cons, and their results will be discussed in the above section.\u003c/p\u003e \u003c/div\u003e"},{"header":"Results and Discussions","content":"\u003cp\u003eOn four different optimizers, including Adam, SGD, RMSProp, and Nadam, this section shows the results of various pre-trained deep learning models, including AlexNet, VGG16, VGG19, ResNet50, Xception, InceptionV3, DenseNet121, MobileNetV1, and EfficientNet B0, on a publicly available dataset, Figshare, for the detection of brain tumors into three categories, glioma, meningioma, and pituitary tumor. Moreover, the performances of the different pre-trained models have been evaluated regarding different statistical performance evaluation parameters, including accuracy [\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e], miss classification rate (MCR) [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e], precision also called positive predicted value (PPV) [\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e], sensitivity also known as recall or true positive rate (TPR) [\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e], specificity also known as true negative rate (TNR) [\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e], F1-score [\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e], intersection over union (IoU) [\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e], negative predictive value (NPV), false negative rate also known as miss rate (FNR), false positive rate also known as fallout (FPR), false discovery rate (FDR), false omission rate (FOR), Matthews correlation coefficient (MCC) [\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e], balanced accuracy, and geometric mean [\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e].\u003cdiv id=\"Equ15\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ15\" name=\"EquationSource\"\u003e\n$$\\:Accuracy=\\:\\frac{TN+TP}{TN+TP+FN+FP}\\:\\times\\:100$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e15\u003c/div\u003e\u003c/div\u003e\u003cdiv id=\"Equ16\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ16\" name=\"EquationSource\"\u003e\n$$\\:Miss\\:classification\\:rate\\:\\left(MCR\\:\\right)=\\:\\frac{FP+FN}{TN+FN+FP+TP}\\:\\times\\:100$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e16\u003c/div\u003e\u003c/div\u003e\u003cdiv id=\"Equ17\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ17\" name=\"EquationSource\"\u003e\n$$\\:Precision=\\:\\frac{TP}{TP+FP}\\:\\times\\:100$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e17\u003c/div\u003e\u003c/div\u003e\u003cdiv id=\"Equ18\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ18\" name=\"EquationSource\"\u003e\n$$\\:Sensitivity=\\:\\frac{TP}{TP+FN}\\:\\times\\:100$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e18\u003c/div\u003e\u003c/div\u003e\u003cdiv id=\"Equ19\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ19\" name=\"EquationSource\"\u003e\n$$\\:Specificity=\\:\\frac{TN}{TN+FP}\\:\\times\\:100$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e19\u003c/div\u003e\u003c/div\u003e\u003cdiv id=\"Equ20\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ20\" name=\"EquationSource\"\u003e\n$$\\:F1-Score=2\\:\\times\\:\\:\\frac{Precision\\:\\times\\:Sensitivity}{Precision+Sensitivity}\\:\\times\\:100$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e20\u003c/div\u003e\u003c/div\u003e\u003cdiv id=\"Equ21\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ21\" name=\"EquationSource\"\u003e\n$$\\:Intersection\\:over\\:Union\\:\\left(IoU\\right)=\\:\\frac{TP}{TP+FP+FN}\\:\\times\\:100$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e21\u003c/div\u003e\u003c/div\u003e\u003cdiv id=\"Equ22\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ22\" name=\"EquationSource\"\u003e\n$$\\:Negative\\:pedictive\\:value\\:\\left(NPV\\right)=\\:\\frac{TN}{TN+FN}\\times\\:100$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e22\u003c/div\u003e\u003c/div\u003e\u003cdiv id=\"Equ23\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ23\" name=\"EquationSource\"\u003e\n$$\\:False\\:negative\\:rate\\:\\left(FNR\\right)=\\:\\frac{FN}{FN+TP}\\:\\times\\:100$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e23\u003c/div\u003e\u003c/div\u003e\u003cdiv id=\"Equ24\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ24\" name=\"EquationSource\"\u003e\n$$\\:False\\:positive\\:rate\\:\\left(FPR\\right)=\\:\\frac{FP}{FP+TN}\\:\\times\\:100$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e24\u003c/div\u003e\u003c/div\u003e\u003cdiv id=\"Equ25\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ25\" name=\"EquationSource\"\u003e\n$$\\:False\\:discovery\\:rate\\:\\left(FDR\\right)=\\:\\frac{FP}{FP+TP}\\:\\times\\:100$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e25\u003c/div\u003e\u003c/div\u003e\u003cdiv id=\"Equ26\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ26\" name=\"EquationSource\"\u003e\n$$\\:False\\:omission\\:rate\\:\\left(FOR\\right)=\\:\\frac{FN}{FN+TN}\\:\\times\\:100$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e26\u003c/div\u003e\u003c/div\u003e\u003cdiv id=\"Equ27\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ27\" name=\"EquationSource\"\u003e\n$$\\:Matthews\\:correlation\\:coefficient\\:\\left(MCC\\right)=\\:\\frac{\\left(TP\\:\\times\\:TN\\right)-(FP\\:\\times\\:FN)}{\\sqrt{(TP+FP)(TP+FN)(TN+FP)(TN+FN)}}\\:\\times\\:100$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e27\u003c/div\u003e\u003c/div\u003e\u003cdiv id=\"Equ28\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ28\" name=\"EquationSource\"\u003e\n$$\\:Balance\\:accuracy=\\:\\frac{Sensitivity+Specificity}{2}\\:\\times\\:100$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e28\u003c/div\u003e\u003c/div\u003e\u003cdiv id=\"Equ29\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ29\" name=\"EquationSource\"\u003e\n$$\\:Geometric\\:mean=\\:\\sqrt{Sensitivity\\:\\times\\:Specificity}\\:\\times\\:100$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e29\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003eTP, or the true positive value, shown in the preceding equations indicates how precisely the model predicts positive classes. TN, the actual negative value, demonstrates how well the model forecasts the negative class. Furthermore, FN stands for the false negative value, which shows the incorrectly projected negative class by the model; FP stands for the false positive, which shows the improperly expected positive class by the model. Based on the above-mentioned values, all the aforementioned statistical performance evaluation criteria were computed; where accuracy is the proportion of correctly identified cases among the total instances. The miss classification rate indicates among all the cases the improperly classified ones. While sensitivity is the proportion of actual positives accurately detected by the model, precision is the proportion of true positive forecasts among all positive predictions. Specificity is the model's proper identification of the actual negatives. The F1-score is the harmonic mean of sensitivity that is, recall and accuracy. The Jaccard index, sometimes known as the IoU, shows how often expected and actual classes overlap. Moreover, NPV shows among all negative forecasts the fraction of real negative predictions. FNR reveals the percentage of true positives mistakenly projected as negatives. While FDR displays the proportion of false positive predictions among all positive predictions, FPR reveals the fraction of real negatives that are mistakenly recognized as positives. FOR shows among all negative predictions the percentage of false negative predictions. Moreover, MCC shows the relationship between expected and actual classes; balance accuracy offers the average of sensitivity and specificity, which is more helpful for an imbalanced dataset; and the geometric mean also shows the geometric mean of sensitivity and specificity, which is likewise helpful for an imbalanced dataset.\u003c/p\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e shows the performance evaluation of the AlexNet model regarding different performance evaluation parameters, including accuracy, MCR, precision, sensitivity, specificity, F1-score, and IoU of each class on different optimizers, while Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e shows the overall performance of the AlexNet model regarding the accuracy, MCR, precision, sensitivity, specificity, F1-score, and IoU, and Table \u003cspan refid=\"Tab5\" class=\"InternalRef\"\u003e5\u003c/span\u003e shows the overall performance of the AlexNet model regarding NPV, FNR, FPR, FDR, FOR, MCC, balance accuracy (BA), and geometric mean (GM) on different optimizers.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003ePerformance of AlexNet model regarding each class on different optimizers.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"9\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOptimizers\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eImage Class\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAccuracy\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eMCR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003ePrecision\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eSensitivity\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eSpecificity\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eF1-Score\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c9\"\u003e \u003cp\u003eIoU\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003e\u003cb\u003eAdam\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGlioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e95.00%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e5.00%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e91.70%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e98.13%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e92.28%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e94.81%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e90.13%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMeningioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e94.78%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e5.22%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e93.62%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e83.02%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e98.31%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e88.00%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e78.57%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePituitary\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e98.04%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.96%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e97.81%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e95.71%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e99.06%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e96.75%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e93.71%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003e\u003cb\u003eSGD\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGlioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e83.26%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e16.74%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e84.42%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e78.50%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e87.40%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e81.36%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e68.57%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMeningioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e82.83%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e17.17%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e63.92%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e58.49%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e90.11%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e61.08%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e43.97%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePituitary\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e94.35%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e5.65%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e84.76%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e99.29%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e92.19%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e91.45%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e84.24%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003e\u003cb\u003eRMSProp\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGlioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e94.13%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e5.87%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e96.06%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e91.12%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e96.75%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e93.53%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e87.84%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMeningioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e91.96%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e8.04%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e89.66%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e73.58%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e97.46%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e80.83%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e67.83%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePituitary\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e93.48%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e6.52%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e82.35%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e100.00%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e90.63%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e90.32%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e82.35%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003e\u003cb\u003eNadam\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGlioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e93.70%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e6.30%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e92.63%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e93.93%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e93.50%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e93.27%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e87.39%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMeningioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e90.43%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e9.57%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e89.74%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e66.04%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e97.74%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e76.09%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e61.40%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePituitary\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e94.57%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e5.43%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e84.85%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e100.00%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e92.19%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e91.80%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e84.85%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e shows the comparison of different evaluation parameters on different optimizers of the AlexNet model regarding each class. The model achieved the best results on Adam Optimizer, with an accuracy of 95%, 94.78%, and 98.04% for glioma, meningioma, and pituitary tumor classes, respectively, with MCRs of 5.0%, 5.22%, and 1.96% for each class, respectively. While the model achieved the lowest results on the SGD optimizer, achieving an accuracy of 83.26%, 82.83%, and 94.35% for glioma, meningioma, and pituitary tumor classes, respectively, with an MCR of 16.74%, 17.17%, and 5.65% for each class respectively.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab4\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 4\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eOverall performance of AlexNet model on different optimizers (Continue).\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"8\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOptimizers\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAccuracy\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eMCR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003ePrecision\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSensitivity\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eSpecificity\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eF1-Score\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eIoU\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eAdam\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e95.94%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e4.06%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e94.38%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e92.29%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e96.55%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e93.19%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e87.47%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eSGD\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e86.81%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e13.19%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e77.70%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e78.76%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e89.90%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e77.96%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e65.60%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eRMSProp\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e93.19%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e6.81%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e89.36%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e88.24%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e94.94%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e88.23%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e79.34%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eNadam\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e92.90%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e7.10%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e89.07%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e86.65%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e94.47%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e87.05%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e77.88%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e shows the overall performance of the AlexNet model on different optimizers regarding accuracy, MCR, precision, sensitivity, specificity, F1-score, and IoU. On all optimizers, Adam Optimizer produced the best results, and the model achieved results such as accuracy of 95.94%, MCR of 4.06%, precision of 94.38%, sensitivity of 92.29%, specificity of 96.55%, F1-score of 93.19%, and IoU of 87.47%. The SGD optimizer produced the worst results, and the model achieved an accuracy of 86.81% with an MCR of 13.19%, precision of 77.70%, sensitivity of 78.76%, specificity of 89.90%, F1-score of 77.96%, and IoU of 65.60%. While RMSProp and Nadam optimizers produced moderate results better than SGD, and the model achieved results on RMSProp with an accuracy of 93.19%, MCR of 6.81%, precision of 89.36%, sensitivity of 88.24%, specificity of 94.94%, F1-score of 88.23%, and IoU of 79.34%. AlexNet achieved results on Nadam Optimizer with an accuracy of 92.90%, MCR of 7.10%, precision of 89.07%, sensitivity of 86.65%, specificity of 94.47%, F1-score of 87.05%, and IoU of 77.88%.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab5\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 5\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eOverall performance of the AlexNet model on different optimizers\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"9\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOptimizers\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNPV\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eFNR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eFPR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eFDR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eFOR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eMCC\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eBA\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c9\"\u003e \u003cp\u003eGM\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eAdam\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e97.16%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e7.71%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e3.45%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e5.62%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e2.84%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e89.20%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e94.42%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e94.29%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eSGD\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e89.97%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e21.24%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e10.10%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e22.30%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e10.03%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e69.03%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e84.33%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e83.70%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eRMSProp\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e95.03%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e11.76%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e5.06%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e10.64%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e4.97%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e83.46%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e91.59%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e91.26%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eNadam\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e95.08%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e13.35%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e5.53%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e10.93%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e4.92%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e81.79%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e90.56%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e90.02%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab5\" class=\"InternalRef\"\u003e5\u003c/span\u003e presents the overall performance of the AlexNet model in terms of NPV, FNR, FDR, FOR, MCC, BA, and GM across different optimizers. As shown, the Adam optimizer produced the best results, while RMSProp and Nadam delivered moderate results, and the SGD optimizer performed the worst. According to the performance metrics in Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e, the model achieved the following values with the Adam optimizer: NPV of 97.16%, FNR of 7.71%, FPR of 3.45%, FDR of 5.62%, FOR of 2.84%, MCC of 89.02%, BA of 94.42%, and GM of 94.29%. In contrast, the model performed less effectively with the SGD optimizer, achieving an NPV of 89.97%, FNR of 21.24%, FPR of 10.10%, FDR of 22.30%, FOR of 10.03%, MCC of 69.03%, BA of 84.33%, and GM of 83.70%. With the RMSProp optimizer, the model attained an NPV of 95.03%, FNR of 11.76%, FPR of 5.06%, FDR of 10.64%, FOR of 4.97%, MCC of 83.46%, and BA and GM values of 91.59% and 91.26%, respectively. Meanwhile, with the Nadam optimizer, the model achieved an NPV of 95.08%, FNR of 13.35%, FPR of 5.53%, FDR of 10.93%, FOR of 4.92%, MCC of 81.79%, and BA and GM values of 90.56% and 90.02%, respectively.\u003c/p\u003e \u003cp\u003eAdditionally, Table\u0026nbsp;\u003cspan refid=\"Tab6\" class=\"InternalRef\"\u003e6\u003c/span\u003e presents the performance evaluation of the VGG16 model across various performance metrics, including accuracy, misclassification rate (MCR), precision, sensitivity, specificity, F1-score, and IoU for each class using different optimizers. Table\u0026nbsp;\u003cspan refid=\"Tab7\" class=\"InternalRef\"\u003e7\u003c/span\u003e displays the overall performance of the VGG16 model in terms of accuracy, MCR, precision, sensitivity, specificity, F1-score, and IoU, while Table\u0026nbsp;\u003cspan refid=\"Tab8\" class=\"InternalRef\"\u003e8\u003c/span\u003e summarizes the overall performance of the VGG16 model based on NPV, FNR, FPR, FDR, FOR, MCC, BA, and GM with different optimizers.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab6\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 6\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003ePerformance of VGG16 model regarding each class on different optimizers.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"9\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOptimizers\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eImage Class\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAccuracy\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eMCR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003ePrecision\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eSensitivity\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eSpecificity\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eF1-Score\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c9\"\u003e \u003cp\u003eIoU\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003e\u003cb\u003eAdam\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGlioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e98.04%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.96%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e97.24%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e98.60%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e97.56%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e97.91%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e95.91%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMeningioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e97.39%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e2.61%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e96.08%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e92.45%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e98.87%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e94.23%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e89.09%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePituitary\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e98.91%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.09%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e97.87%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e98.57%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e99.06%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e98.22%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e96.50%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003e\u003cb\u003eSGD\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGlioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e95.22%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e4.78%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e94.44%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e95.33%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e95.12%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e94.88%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e90.27%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMeningioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e92.61%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e7.39%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e86.00%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e81.13%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e96.05%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e83.50%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e71.67%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePituitary\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e97.39%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e2.61%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e94.44%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e97.14%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e97.50%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e95.77%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e91.89%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003e\u003cb\u003eRMSProp\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGlioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e98.04%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.96%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e97.67%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e98.13%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e97.97%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e97.90%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e95.89%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMeningioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e97.39%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e2.61%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e97.96%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e90.57%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e99.44%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e94.12%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e88.89%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePituitary\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e98.04%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.96%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e94.56%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e99.29%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e97.50%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e96.86%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e93.92%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003e\u003cb\u003eNadam\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGlioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e98.70%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.30%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e99.52%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e97.66%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e99.59%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e98.58%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e97.21%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMeningioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e97.83%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e2.17%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e93.64%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e97.17%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e98.02%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e95.37%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e91.15%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePituitary\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e98.70%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.30%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e97.86%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e97.86%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e99.06%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e97.86%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e95.80%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab6\" class=\"InternalRef\"\u003e6\u003c/span\u003e shows the comparison of different evaluation parameters on different optimizers of the VGG16 model regarding each class. The model achieved the best results on Nadam Optimizer, with an accuracy of 98.70%, 97.38%, and 98.70% for glioma, meningioma, and pituitary tumor classes, respectively, with MCRs of 1.30%, 2.17%, and 1.30% for each class, respectively. While the model achieved the lowest results on the SGD optimizer, achieving an accuracy of 95.22%, 92.61%, and 97.39% for glioma, meningioma, and pituitary tumor classes, respectively, with an MCR of 4.78%, 7.39%, and 2.61% for each class respectively.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab7\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 7\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eOverall performance of VGG16 model on different optimizers (Continue).\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"8\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOptimizers\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAccuracy\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eMCR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003ePrecision\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSensitivity\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eSpecificity\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eF1-Score\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eIoU\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eAdam\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e98.12%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.88%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e97.06%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e96.54%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e98.50%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e96.79%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e93.83%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eSGD\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e95.07%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e4.93%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e91.63%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e91.20%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e96.22%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e91.38%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e84.61%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eRMSProp\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e97.83%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e2.17%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e96.73%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e95.99%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e98.30%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e96.29%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e92.90%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eNadam\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e98.41%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.59%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e97.01%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e97.56%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e98.89%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e97.27%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e94.72%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab7\" class=\"InternalRef\"\u003e7\u003c/span\u003e presents the overall performance of the VGG16 model using different optimizers in terms of accuracy, MCR, precision, sensitivity, specificity, F1-score, and IoU. Among all optimizers, the Nadam optimizer produced the best results, with the model achieving an accuracy of 98.41%, an MCR of 1.59%, a precision of 97.01%, a sensitivity of 97.56%, a specificity of 98.89%, an F1-score of 97.27%, and an IoU of 94.72%. In contrast, the SGD optimizer yielded the poorest performance, with the model achieving an accuracy of 95.07%, an MCR of 4.93%, a precision of 91.63%, a sensitivity of 91.20%, a specificity of 96.22%, an F1-score of 91.38%, and an IoU of 84.61%. The RMSProp and Adam optimizers produced moderate results, better than SGD. Using RMSProp, the model achieved an accuracy of 97.83%, an MCR of 2.17%, a precision of 96.73%, a sensitivity of 95.99%, a specificity of 98.30%, an F1-score of 96.29%, and an IoU of 92.90%. With the Adam optimizer, VGG16 achieved an accuracy of 98.12%, an MCR of 1.88%, a precision of 97.06%, a sensitivity of 96.54%, a specificity of 98.50%, an F1-score of 96.79%, and an IoU of 93.83%.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab8\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 8\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eOverall performance of VGG16 model on different optimizers\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"9\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOptimizers\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNPV\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eFNR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eFPR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eFDR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eFOR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eMCC\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eBA\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c9\"\u003e \u003cp\u003eGM\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eAdam\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e98.63%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e3.46%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.50%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e2.94%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1.37%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e95.12%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e97.52%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e97.50%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eSGD\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e96.36%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e8.80%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e3.78%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e8.37%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e3.64%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e87.53%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e93.71%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e93.61%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eRMSProp\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e98.43%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e4.01%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.70%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e3.27%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1.57%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e94.39%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e97.15%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e97.11%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eNadam\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e98.74%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e2.44%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.11%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e2.99%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1.26%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e96.38%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e98.23%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e98.23%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab8\" class=\"InternalRef\"\u003e8\u003c/span\u003e presents the overall performance of the VGG16 model in terms of NPV, FNR, FDR, FOR, MCC, BA, and GM across different optimizers. As observed, the Nadam optimizer produced the best results, while RMSProp and Adam yielded moderate results, and the SGD optimizer produced the lowest performance. Based on the performance parameters from Table\u0026nbsp;\u003cspan refid=\"Tab6\" class=\"InternalRef\"\u003e6\u003c/span\u003e, the model achieved the following values using the Adam optimizer: NPV of 98.63%, FNR of 3.46%, FPR of 1.50%, FDR of 2.94%, FOR of 1.37%, MCC of 95.12%, BA of 97.52%, and GM of 97.50%. Conversely, with the SGD optimizer, the model showed reduced performance with an NPV of 96.36%, FNR of 8.80%, FPR of 3.78%, FDR of 8.37%, FOR of 3.64%, MCC of 87.53%, BA of 93.71%, and GM of 93.61%. Additionally, with the RMSProp optimizer, the model achieved an NPV of 98.43%, FNR of 4.01%, FPR of 1.70%, FDR of 3.27%, FOR of 1.57%, MCC of 94.39%, and BA and GM of 97.15% and 97.11%, respectively. The Nadam optimizer delivered the best results with an NPV of 98.74%, FNR of 2.44%, FPR of 1.11%, FDR of 2.99%, FOR of 1.26%, MCC of 96.38%, and BA and GM of 98.23% and 98.23%, respectively.\u003c/p\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab9\" class=\"InternalRef\"\u003e9\u003c/span\u003e provides the performance evaluation of the VGG19 model based on various parameters, including accuracy, MCR, precision, sensitivity, specificity, F1-score, and IoU for each class across different optimizers. Table\u0026nbsp;\u003cspan refid=\"Tab10\" class=\"InternalRef\"\u003e10\u003c/span\u003e summarizes the overall performance of the VGG19 model regarding accuracy, MCR, precision, sensitivity, specificity, F1-score, and IoU. Finally, Table\u0026nbsp;\u003cspan refid=\"Tab11\" class=\"InternalRef\"\u003e11\u003c/span\u003e presents the overall performance of the VGG19 model in terms of NPV, FNR, FPR, FDR, FOR, MCC, balanced accuracy (BA), and geometric mean (GM) across different optimizers.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab9\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 9\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003ePerformance of VGG19 model regarding each class on different optimizers.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"9\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOptimizers\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eImage Class\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAccuracy\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eMCR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003ePrecision\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eSensitivity\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eSpecificity\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eF1-Score\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c9\"\u003e \u003cp\u003eI0U\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003e\u003cb\u003eAdam\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGlioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e99.13%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.87%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e99.07%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e99.07%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e99.19%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e99.07%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e98.15%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMeningioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e98.70%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.30%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e98.08%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e96.23%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e99.44%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e97.14%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e94.44%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePituitary\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e99.57%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.43%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e98.59%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e100.00%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e99.38%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e99.29%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e98.59%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003e\u003cb\u003eSGD\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGlioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e95.22%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e4.78%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e93.24%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e96.73%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e93.90%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e94.95%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e90.39%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMeningioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e92.61%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e7.39%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e91.86%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e74.53%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e98.02%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e82.29%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e69.91%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePituitary\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e96.52%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e3.48%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e90.79%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e98.57%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e95.63%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e94.52%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e89.61%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003e\u003cb\u003eRMSProp\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGlioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e98.91%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.09%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e99.06%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e98.60%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e99.19%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e98.83%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e97.69%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMeningioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e97.83%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e2.17%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e97.06%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e93.40%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e99.15%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e95.19%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e90.83%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePituitary\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e98.91%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.09%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e96.55%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e100.00%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e98.44%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e98.25%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e96.55%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003e\u003cb\u003eNadam\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGlioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e98.26%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.74%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e98.13%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e98.13%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e98.37%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e98.13%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e96.33%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMeningioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e98.04%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.96%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e96.19%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e95.28%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e98.87%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e95.73%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e91.82%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePituitary\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e99.35%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.65%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e98.58%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e99.29%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e99.38%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e98.93%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e97.89%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab9\" class=\"InternalRef\"\u003e9\u003c/span\u003e shows the comparison of different evaluation parameters on different optimizers of the VGG19 model regarding each class. The model achieved the best results on Adam Optimizer, with an accuracy of 99.13%, 98.70%, and 99.57% for glioma, meningioma, and pituitary tumor classes, respectively, with MCRs of 0.87%, 1.30%, and 0.43% for each class, respectively. While the model achieved the lowest results on the SGD optimizer, achieving accuracy of 95.22%, 92.61%, and 96.52% for glioma, meningioma, and pituitary tumor classes, respectively, with MCRs of 4.78%, 7.39%, and 3.48% for each class respectively.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab10\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 10\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eOverall performance of VGG19 model on different optimizers (Continue).\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"8\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOptimizers\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAccuracy\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eMCR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003ePrecision\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSensitivity\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eSpecificity\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eF1-Score\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eI0U\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eAdam\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e99.13%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.87%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e98.58%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e98.43%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e99.33%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e98.50%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e97.06%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eSGD\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e94.78%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e5.22%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e91.96%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e89.94%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e95.85%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e90.59%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e83.30%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eRMSProp\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e98.55%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.45%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e97.56%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e97.33%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e98.93%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e97.42%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e95.02%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eNadam\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e98.55%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.45%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e97.63%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e97.57%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e98.87%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e97.60%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e95.35%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab10\" class=\"InternalRef\"\u003e10\u003c/span\u003e shows the overall performance of the VGG19 model on different optimizers regarding accuracy, MCR, precision, sensitivity, specificity, F1-score, and IoU. On all optimizers, Adam Optimizer produced the best results, and the model achieved results such as accuracy of 99.13%, MCR of 0.87%, precision of 98.58%, sensitivity of 98.43%, specificity of 99.33%, F1-score of 98.50%, and IoU of 97.06%. The SGD optimizer produced the worst results, and the model achieved an accuracy of 94.78% with an MCR of 5.22%, precision of 91.96%, sensitivity of 89.94%, specificity of 95.85%, F1-score of 90.59%, and IoU of 83.30%. While RMSProp and Nadam optimizers produced moderate results better than SGD, and the model achieved results on RMSProp with an accuracy of 98.55%, MCR of 1.45%, precision of 97.56%, sensitivity of 97.33%, specificity of 98.93%, F1-score of 97.42%, and IoU of 95.02%. while VGG19 achieved results on Nadam Optimizer with an accuracy of 98.55%, MCR of 1.45%, precision of 97.63%, sensitivity of 97.57%, specificity of 98.87%, F1-score of 97.60%, and IoU of 95.35%.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab11\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 11\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eOverall performance of the VGG19 model on different optimizers.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"9\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOptimizers\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNPV\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eFNR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eFPR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eFDR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eFOR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eMCC\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eBA\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c9\"\u003e \u003cp\u003eGM\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eAdam\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e99.35%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.57%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.67%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.42%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.65%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e97.78%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e98.88%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e98.88%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eSGD\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e96.40%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e10.06%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e4.15%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e8.04%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e3.60%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e86.25%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e92.90%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e92.62%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eRMSProp\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e98.94%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e2.67%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.07%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e2.44%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1.06%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e96.28%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e98.13%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e98.11%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eNadam\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e98.88%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e2.43%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.13%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e2.37%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1.12%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e96.45%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e98.22%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e98.21%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab11\" class=\"InternalRef\"\u003e11\u003c/span\u003e shows the overall performance of the VGG19 model regarding NPV, FNR, FDR, FOR, MCC, BA, and GM on different optimizers. As we see above, Adam Optimizer produced better results, RMSProp and Nadam produced moderate results, and SGD Optimizer produced the worst results. Regarding Table \u003cspan refid=\"Tab9\" class=\"InternalRef\"\u003e9\u003c/span\u003e performance parameters, the model achieved values on Adam Optimizer as NPV of 99.35%, FNR of 1.57%, FPR of 0.67%, FDR of 1.42%, FOR of 0.65%, MCC of 97.78%, BA of 98.88%, and GM of 98.88%. While on the SGD optimizer, the model achieved less results, as NPV of 96.40%, FNR of 10.06%, FPR of 4.15%, FDR of 8.04%, FOR of 3.60%, an MCC of 86.25%, a BA of 92.90%, and a GM of 92.62%. Further, the RMSProp model achieved a NPV of 98.94%, a FNR of 2.67%, FPR of 1.07%, FDR of 2.44%, FOR of 1.06%, MCC of 96.28%, BA and GM of 98.13%, and 98.11%, respectively. While on Nadam Optimizer, the model achieved the best results, as NPV of 98.88%, FNR of 2.43%, FPR of 1.13%, FDR of 2.37%, FOR of 1.12%, MCC of 96.45%, BA and GM of 98.22% and 98.21%, respectively.\u003c/p\u003e \u003cp\u003eAdditionally, Table \u003cspan refid=\"Tab12\" class=\"InternalRef\"\u003e12\u003c/span\u003e shows the performance evaluation of the ResNet50 model regarding different performance evaluation parameters, including accuracy, MCR, precision, sensitivity, specificity, F1-score, and IoU of each class on different optimizers, while Table\u0026nbsp;\u003cspan refid=\"Tab13\" class=\"InternalRef\"\u003e13\u003c/span\u003e shows the overall performance of the ResNet50 model regarding the accuracy, MCR, precision, sensitivity, specificity, F1-score, and IoU, and Table \u003cspan refid=\"Tab14\" class=\"InternalRef\"\u003e14\u003c/span\u003e shows the overall performance of the ResNet50 model regarding NPV, FNR, FPR, FDR, FOR, MCC, BA, and GM on different optimizers.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab12\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 12\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003ePerformance of ResNet50 model regarding each class on different optimizers.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"9\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOptimizers\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eImage Class\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAccuracy\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eMCR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003ePrecision\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eSensitivity\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eSpecificity\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eF1-Score\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c9\"\u003e \u003cp\u003eIoU\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003e\u003cb\u003eAdam\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGlioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e98.91%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.09%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e99.06%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e98.60%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e99.19%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e98.83%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e97.69%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMeningioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e97.61%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e2.39%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e97.03%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e92.45%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e99.15%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e94.69%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e89.91%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePituitary\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e98.70%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.30%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e95.89%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e100.00%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e98.13%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e97.90%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e95.89%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003e\u003cb\u003eSGD\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGlioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e91.96%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e8.04%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e90.41%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e92.52%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e91.46%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e91.45%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e84.26%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMeningioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e87.17%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e12.83%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e78.31%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e61.32%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e94.92%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e68.78%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e52.42%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePituitary\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e93.48%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e6.52%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e84.81%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e95.71%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e92.50%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e89.93%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e81.71%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003e\u003cb\u003eRMSProp\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGlioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e99.13%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.87%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e99.07%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e99.07%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e99.19%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e99.07%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e98.15%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMeningioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e97.61%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e2.39%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e98.97%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e90.57%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e99.72%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e94.58%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e89.72%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePituitary\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e98.04%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.96%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e93.96%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e100.00%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e97.19%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e96.89%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e93.96%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003e\u003cb\u003eNadam\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGlioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e99.13%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.87%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e98.61%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e99.53%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e98.78%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e99.07%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e98.16%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMeningioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e96.30%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e3.70%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e98.90%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e84.91%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e99.72%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e91.37%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e84.11%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePituitary\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e97.17%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e2.83%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e91.50%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e100.00%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e95.94%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e95.56%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e91.50%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab12\" class=\"InternalRef\"\u003e12\u003c/span\u003e shows the comparison of different evaluation parameters on different optimizers of the ResNet50 model regarding each class. The model achieved the best results on Adam Optimizer, with an accuracy of 98.91%, 97.61%, and 98.70% for glioma, meningioma, and pituitary tumor classes, respectively, with MCRs of 1.09%, 2.39%, and 1.30% for each class, respectively. While the model achieved the lowest results on the SGD optimizer, achieving an accuracy of 91.96%, 87.17%, and 93.48% for glioma, meningioma, and pituitary tumor classes, respectively, with MCRs of 8.04%, 12.83%, and 6.52% for each class respectively.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab13\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 13\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eOverall performance of ResNet50 model on different optimizers (Continue).\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"8\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOptimizers\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAccuracy\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eMCR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003ePrecision\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSensitivity\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eSpecificity\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eF1-Score\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eIoU\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eAdam\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e98.41%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.59%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e97.33%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e97.02%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e98.82%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e97.14%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e94.49%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eSGD\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e90.87%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e9.13%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e84.51%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e83.19%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e92.96%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e83.39%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e72.79%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eRMSProp\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e98.26%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.74%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e97.33%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e96.54%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e98.70%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e96.84%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e93.94%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eNadam\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e97.54%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e2.46%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e96.34%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e94.81%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e98.15%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e95.33%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e91.26%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab13\" class=\"InternalRef\"\u003e13\u003c/span\u003e shows the overall performance of the ResNet50 model on different optimizers regarding accuracy, MCR, precision, sensitivity, specificity, F1-score, and IoU. On all optimizers, Adam Optimizer produced the best results, and the model achieved results such as accuracy of 98.41%, MCR of 1.59%, precision of 97.33%, sensitivity of 97.02%, specificity of 98.82%, F1-score of 97.14%, and IoU of 94.49%. The SGD optimizer produced the worst results, and the model achieved an accuracy of 90.87% with MCR of 9.13%, precision of 84.51%, sensitivity of 83.19%, specificity of 92.96%, F1-score of 83.39%, and IoU of 72.79%. While RMSProp and Nadam optimizers produced moderate results better than SGD, and the model achieved results on RMSProp as accuracy of 98.26%, MCR of 1.74%, precision of 97.33%, sensitivity of 96.54%, specificity of 98.70%, F1-score of 96.84%, and IoU of 93.94%. while ResNet50 achieved results on Nadam Optimizer as accuracy of 97.54%, MCR of 2.46%, precision of 96.34%, sensitivity of 94.81%, specificity of 98.15%, F1-score of 95.33%, and IoU of 91.26%.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab14\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 14\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eOverall performance of the ResNet50 model on different optimizers\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"9\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOptimizers\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNPV\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eFNR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eFPR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eFDR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eFOR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eMCC\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eBA\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c9\"\u003e \u003cp\u003eGM\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eAdam\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e98.85%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e2.98%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.18%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e2.67%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1.15%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e95.87%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e97.92%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e97.90%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eSGD\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e93.50%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e16.81%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e7.04%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e15.49%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e6.50%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e76.72%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e88.07%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e87.46%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eRMSProp\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e98.81%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e3.46%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.30%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e2.67%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1.19%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e95.33%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e97.62%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e97.58%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eNadam\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e98.42%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e5.19%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.85%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e3.66%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1.58%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e93.18%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e96.48%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e96.37%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab14\" class=\"InternalRef\"\u003e14\u003c/span\u003e shows the overall performance of the ResNet50 model regarding NPV, FNR, FDR, FOR, MCC, BA, and GM on different optimizers. As we see above, Adam Optimizer produced better results, RMSProp and Nadam produced moderate results, and SGD Optimizer produced the worst results. Regarding Table \u003cspan refid=\"Tab12\" class=\"InternalRef\"\u003e12\u003c/span\u003e performance parameters, the model achieved values on Adam Optimizer as NPV of 98.85%, FNR of 2.98%, FPR of 1.18%, FDR of 2.67%, FOR of 1.15%, MCC of 95.87%, BA of 97.92%, and GM of 97.90%. While on the SGD optimizer, the model achieved less results, as NPV of 93.50%, FNR of 16.81%, FPR of 7.04%, FDR of 15.49%, FOR of 6.50%, a MCC of 76.72%, a BA of 88.07%, and a GM of 87.46%. Further, the RMSProp model achieved a NPV of 98.81%, a FNR of 3.46%, FPR of 1.30%, FDR of 2.67%, FOR of 1.19%, MCC of 95.33%, BA and GM of 97.62%, and 97.58%, respectively. While on Nadam Optimizer, the model achieved results as NPV of 98.42%, FNR of 5.19%, FPR of 1.85%, FDR of 3.66%, FOR of 1.58%, MCC of 93.18%, BA and GM of 96.48% and 96.37%, respectively.\u003c/p\u003e \u003cp\u003eMoreover, Table \u003cspan refid=\"Tab15\" class=\"InternalRef\"\u003e15\u003c/span\u003e shows the performance evaluation of the Xception model regarding different performance evaluation parameters, including accuracy, MCR, precision, sensitivity, specificity, F1-score, and IoU of each class on different optimizers, while Table\u0026nbsp;\u003cspan refid=\"Tab16\" class=\"InternalRef\"\u003e16\u003c/span\u003e shows the overall performance of the Xception model regarding accuracy, MCR, precision, sensitivity, specificity, F1-score, and IoU, and Table \u003cspan refid=\"Tab17\" class=\"InternalRef\"\u003e17\u003c/span\u003e shows the overall performance of the Xception model regarding NPV, FNR, FPR, FDR, FOR, MCC, BA, and GM on different optimizers.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab15\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 15\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003ePerformance of the Xception model regarding each class on different optimizers.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"9\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOptimizers\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eImage Class\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAccuracy\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eMCR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003ePrecision\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eSensitivity\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eSpecificity\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eF1-Score\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c9\"\u003e \u003cp\u003eIoU\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003e\u003cb\u003eAdam\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGlioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e99.35%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.65%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e100.00%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e98.60%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e100.00%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e99.29%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e98.60%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMeningioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e97.83%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e2.17%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e96.15%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e94.34%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e98.87%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e95.24%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e90.91%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePituitary\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e98.48%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.52%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e95.86%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e99.29%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e98.13%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e97.54%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e95.21%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003e\u003cb\u003eSGD\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGlioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e56.74%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e43.26%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e51.86%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e97.66%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e21.14%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e67.75%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e51.23%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMeningioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e77.39%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e22.61%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e66.67%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e3.77%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e99.44%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e7.14%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e3.70%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePituitary\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e74.13%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e25.87%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e70.59%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e25.71%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e95.31%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e37.70%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e23.23%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003e\u003cb\u003eRMSProp\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGlioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e99.78%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.22%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e100.00%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e99.53%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e100.00%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e99.77%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e99.53%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMeningioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e98.48%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.52%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e98.06%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e95.28%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e99.44%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e96.65%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e93.52%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePituitary\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e98.70%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.30%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e96.53%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e99.29%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e98.44%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e97.89%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e95.86%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003e\u003cb\u003eNadam\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGlioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e99.35%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.65%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e99.53%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e99.07%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e99.59%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e99.30%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e98.60%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMeningioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e97.83%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e2.17%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e97.06%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e93.40%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e99.15%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e95.19%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e90.83%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePituitary\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e98.48%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.52%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e95.86%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e99.29%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e98.13%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e97.54%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e95.21%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab15\" class=\"InternalRef\"\u003e15\u003c/span\u003e shows the comparison of different evaluation parameters on different optimizers of the Xception model regarding each class. The model achieved the best results on RMSProp Optimizer, with an accuracy of 99.78%, 98.48%, and 98.70% for glioma, meningioma, and pituitary tumor classes, respectively, with MCRs of 0.22%, 1.52%, and 1.30% for each class, respectively. While the model achieved the lowest results on the SGD optimizer, achieving an accuracy of 56.74%, 77.39%, and 74.13% for glioma, meningioma, and pituitary tumor classes, respectively, with MCRs of 43.26%, 22.61%, and 25.87% for each class respectively.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab16\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 16\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eOverall performance of the Xception model on different optimizers (Continue).\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"8\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOptimizers\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAccuracy\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eMCR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003ePrecision\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSensitivity\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eSpecificity\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eF1-Score\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eIoU\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eAdam\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e98.55%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.45%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e97.34%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e97.41%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e99.00%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e97.36%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e94.90%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eSGD\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e69.42%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e30.58%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e63.04%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e42.38%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e71.96%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e37.53%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e26.05%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eRMSProp\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e98.99%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.01%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e98.20%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e98.03%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e99.29%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e98.10%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e96.30%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eNadam\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e98.55%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.45%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e97.48%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e97.25%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e98.96%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e97.34%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e94.88%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab16\" class=\"InternalRef\"\u003e16\u003c/span\u003e shows the overall performance of the Xception model on different optimizers regarding accuracy, MCR, precision, sensitivity, specificity, F1-score, and IoU. On all optimizers, RMSProp Optimizer produced the best results, and the model achieved results such as accuracy of 98.99%, MCR of 1.01%, precision of 98.20%, sensitivity of 98.03%, specificity of 99.29%, F1-score of 98.10%, and IoU of 96.30%. The SGD optimizer produced the worst results, and the model achieved an accuracy of 69.42% with MCR of 30.58%, precision of 63.04%, sensitivity of 42.38%, specificity of 71.96%, F1-score of 37.53%, and IoU of 26.05%. While Adam and Nadam optimizers produced moderate results better than SGD, and model achieved results on Adam as accuracy of 98.55%, MCR of 1.45%, precision of 97.34%, sensitivity of 97.41%, specificity of 99.00%, F1-score of 97.36%, and IoU of 94.90%. While Xception achieved results on Nadam Optimizer as accuracy of 98.55%, MCR of 1.45%, precision of 97.48%, sensitivity of 97.25%, specificity of 98.96%, F1-score of 97.34%, and IoU of 94.88%.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab17\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 17\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eOverall performance of the Xception model on different optimizers.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"9\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOptimizers\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNPV\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eFNR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eFPR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eFDR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eFOR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eMCC\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eBA\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c9\"\u003e \u003cp\u003eGM\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eAdam\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e98.93%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e2.59%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.00%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e2.66%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1.07%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e96.38%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e98.20%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e98.19%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eSGD\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e81.11%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e57.62%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e28.04%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e36.96%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e18.89%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e20.16%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e57.17%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e38.10%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eRMSProp\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e99.29%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.97%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.71%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.80%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.71%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e97.33%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e98.66%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e98.65%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eNadam\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e98.97%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e2.75%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.04%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e2.52%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1.03%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e96.22%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e98.10%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e98.09%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab17\" class=\"InternalRef\"\u003e17\u003c/span\u003e shows the overall performance of the Xception model regarding NPV, FNR, FDR, FOR, MCC, BA, and GM on different optimizers. As we see above, RMSProp Optimizer produced better results, Adam and Nadam produced moderate results, and SGD Optimizer produced the worst results. Regarding Table \u003cspan refid=\"Tab15\" class=\"InternalRef\"\u003e15\u003c/span\u003e performance parameters, the model achieved values on Adam Optimizer as NPV of 98.93%, FNR of 2.59%, FPR of 1.00%, FDR of 2.66%, FOR of 1.07%, MCC of 96.38%, BA of 98.20%, and GM of 98.19%. While on the SGD optimizer, the model achieved less results, as NPV of 81.11%, FNR of 57.62%, FPR of 28.04%, FDR of 36.96%, FOR of 18.89%, a MCC of 20.16%, a BA of 57.17%, and a GM of 38.10%. Further, the RMSProp model achieved a NPV of 99.29%, a FNR of 1.97%, FPR of 0.71%, FDR of 1.80%, FOR of 0.71%, MCC of 97.33%, BA and GM of 98.66%, and 98.65%, respectively. While on Nadam Optimizer, the model achieved results as NPV of 98.97%, FNR of 2.75%, FPR of 1.04%, FDR of 2.52%, FOR of 1.03%, MCC of 96.22%, BA and GM of 98.10% and 98.09%, respectively.\u003c/p\u003e \u003cp\u003eBelow Table \u003cspan refid=\"Tab18\" class=\"InternalRef\"\u003e18\u003c/span\u003e shows the performance evaluation of the InceptionV3 model regarding different performance evaluation parameters, including accuracy, MCR, precision, sensitivity, specificity, F1-score, and IoU of each class on different optimizers. In contrast, Table\u0026nbsp;\u003cspan refid=\"Tab19\" class=\"InternalRef\"\u003e19\u003c/span\u003e shows the overall performance of the InceptionV3 model regarding accuracy, MCR, precision, sensitivity, specificity, F1-score, and IoU, and Table\u0026nbsp;\u003cspan refid=\"Tab20\" class=\"InternalRef\"\u003e20\u003c/span\u003e shows the overall performance of the InceptionV3 model regarding NPV, FNR, FPR, FDR, FOR, MCC, BA, and (GM) on different optimizers.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab18\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 18\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003ePerformance of InceptionV3 model regarding each class on different optimizers.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"9\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOptimizers\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eImage Class\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAccuracy\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eMCR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003ePrecision\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eSensitivity\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eSpecificity\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eF1-Score\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c9\"\u003e \u003cp\u003eI0U\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003e\u003cb\u003eAdam\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGlioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e99.35%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.65%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e99.07%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e99.53%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e99.19%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e99.30%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e98.61%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMeningioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e97.83%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e2.17%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e98.00%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e92.45%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e99.44%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e95.15%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e90.74%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePituitary\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e98.48%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.52%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e95.86%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e99.29%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e98.13%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e97.54%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e95.21%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003e\u003cb\u003eSGD\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGlioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e83.70%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e16.30%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e75.65%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e95.79%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e73.17%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e84.54%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e73.21%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMeningioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e81.30%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e18.70%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e76.32%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e27.36%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e97.46%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e40.28%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e25.22%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePituitary\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e88.48%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e11.52%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e78.81%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e85.00%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e90.00%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e81.79%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e69.19%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003e\u003cb\u003eRMSProp\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGlioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e99.35%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.65%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e99.53%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e99.07%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e99.59%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e99.30%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e98.60%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMeningioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e98.26%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.74%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e97.12%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e95.28%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e99.15%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e96.19%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e92.66%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePituitary\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e98.91%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.09%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e97.20%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e99.29%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e98.75%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e98.23%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e96.53%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003e\u003cb\u003eNadam\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGlioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e99.78%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.22%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e100.00%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e99.53%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e100.00%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e99.77%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e99.53%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMeningioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e98.04%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.96%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e97.09%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e94.34%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e99.15%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e95.69%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e91.74%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePituitary\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e98.26%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.74%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e95.83%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e98.57%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e98.13%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e97.18%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e94.52%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab18\" class=\"InternalRef\"\u003e18\u003c/span\u003e shows the comparison of different evaluation parameters on different optimizers of the InceptionV3 model regarding each class. The model achieved the best results on RMSProp Optimizer, with an accuracy of 99.35%, 98.26%, and 98.91% for glioma, meningioma, and pituitary tumor classes, respectively, with MCRs of 0.65%, 1.74%, and 1.09% for each class, respectively. While the model achieved the lowest results on the SGD optimizer, achieving accuracy of 83.70%, 81.30%, and 88.48% for glioma, meningioma, and pituitary tumor classes, respectively, with MCRs of 16.30%, 18.70%, and 11.52% for each class respectively.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab19\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 19\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eOverall performance of InceptionV3 model on different optimizers (Continue).\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"8\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOptimizers\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAccuracy\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eMCR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003ePrecision\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSensitivity\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eSpecificity\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eF1-Score\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eIoU\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eAdam\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e98.55%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.45%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e97.64%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e97.09%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e98.92%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e97.33%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e94.85%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eSGD\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e84.49%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e15.51%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e76.92%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e69.38%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e86.88%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e68.87%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e55.87%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eRMSProp\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e98.84%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.16%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e97.95%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e97.88%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e99.17%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e97.91%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e95.93%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eNadam\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e98.70%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.30%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e97.64%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e97.48%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e99.09%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e97.55%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e95.27%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab19\" class=\"InternalRef\"\u003e19\u003c/span\u003e shows the overall performance of the InceptionV3 model on different optimizers regarding accuracy, MCR, precision, sensitivity, specificity, F1-score, and IoU. On all optimizers, RMSProp Optimizer produced the best results, and the model achieved results such as accuracy of 98.84%, MCR of 1.16%, precision of 97.95%, sensitivity of 97.88%, specificity of 99.17%, F1-score of 97.91%, and IoU of 95.93%. While the SGD optimizer produced the worst results, and the model achieved an accuracy of 84.49% with an MCR of 15.51%, precision of 76.92%, sensitivity of 69.38%, specificity of 86.88%, F1-score of 68.87%, and IoU of 55.87%. While Adam and Nadam optimizers produced moderate results better than SGD and model achieved results on Adam with an accuracy of 98.55%, MCR of 1.45%, precision of 97.64%, sensitivity of 97.09%, specificity of 98.92%, F1-score of 97.33%, and IoU of 94.85%. while InceptionV3 achieved results on Nadam Optimizer as accuracy of 98.70%, MCR of 1.30%, precision of 97.64%, sensitivity of 97.48%, specificity of 99.09%, F1-score of 97.55%, and IoU of 95.27%.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab20\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 20\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eOverall performance of the InceptionV3 model on different optimizers.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"9\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOptimizers\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNPV\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eFNR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eFPR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eFDR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eFOR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eMCC\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eBA\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c9\"\u003e \u003cp\u003eGM\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eAdam\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e99.02%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e2.91%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.08%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e2.36%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.98%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e96.07%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e98.00%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e97.98%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eSGD\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e90.07%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e30.62%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e13.12%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e23.08%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e9.93%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e59.24%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e78.13%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e74.27%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eRMSProp\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e99.16%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e2.12%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.83%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e2.05%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.84%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e97.04%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e98.52%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e98.52%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eNadam\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e99.09%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e2.52%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.91%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e2.36%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.91%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e96.58%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e98.29%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e98.28%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab20\" class=\"InternalRef\"\u003e20\u003c/span\u003e shows the overall performance of the InceptionV3 model regarding NPV, FNR, FDR, FOR, MCC, BA, and GM on different optimizers. As we see above, RMSProp Optimizer produced better results, Adam and Nadam produced moderate results, and SGD Optimizer produced the worst results. Regarding Table \u003cspan refid=\"Tab18\" class=\"InternalRef\"\u003e18\u003c/span\u003e performance parameters, the model achieved values on Adam Optimizer as NPV of 99.02%, FNR of 2.91%, FPR of 1.08%, FDR of 2.36%, FOR of 0.98%, MCC of 96.07%, BA of 98.00%, and GM of 97.98%. While on the SGD optimizer, the model achieved less results, as NPV of 90.07%, FNR of 30.62%, FPR of 13.12%, FDR of 23.08%, FOR of 9.93%, a MCC of 59.24%, a BA of 78.13%, and a GM of 74.27%. Further, the RMSProp model achieved a NPV of 99.16%, a FNR of 2.12%, FPR of 0.83%, FDR of 2.05%, FOR of 0.84%, MCC of 97.04%, BA and GM of 98.52%, and 98.52%, respectively. While on Nadam Optimizer, the model achieved results as NPV of 99.09%, FNR of 2.52%, FPR of 0.91%, FDR of 2.36%, FOR of 0.91%, MCC of 96.58%, BA and GM of 98.29% and 98.28%, respectively.\u003c/p\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab21\" class=\"InternalRef\"\u003e21\u003c/span\u003e shows the performance evaluation of the MobileNetV1 model regarding different performance evaluation parameters, including accuracy, MCR, precision, sensitivity, specificity, F1-score, and IoU of each class on different optimizers, while Table\u0026nbsp;\u003cspan refid=\"Tab22\" class=\"InternalRef\"\u003e22\u003c/span\u003e shows the overall performance of the MobileNetV1 model regarding accuracy, MCR, precision, sensitivity, specificity, F1-score, and IoU, and Table \u003cspan refid=\"Tab23\" class=\"InternalRef\"\u003e23\u003c/span\u003e shows the overall performance of the MobileNetV1 model regarding NPV, FNR, FPR, FDR, FOR, MCC, BA, and GM on different optimizers.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab21\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 21\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003ePerformance of MobileNetV1 model regarding each class on different optimizers.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"9\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOptimizers\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eImage Class\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAccuracy\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eMCR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003ePrecision\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eSensitivity\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eSpecificity\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eF1-Score\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c9\"\u003e \u003cp\u003eIoU\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003e\u003cb\u003eAdam\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGlioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e99.35%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.65%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e99.07%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e99.53%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e99.19%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e99.30%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e98.61%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMeningioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e98.04%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.96%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e98.99%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e92.45%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e99.72%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e95.61%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e91.59%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePituitary\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e98.70%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.30%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e95.89%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e100.00%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e98.13%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e97.90%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e95.89%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003e\u003cb\u003eSGD\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGlioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e91.30%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e8.70%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e91.43%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e89.72%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e92.68%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e90.57%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e82.76%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMeningioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e87.17%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e12.83%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e74.74%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e66.98%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e93.22%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e70.65%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e54.62%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePituitary\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e93.70%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e6.30%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e85.81%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e95.00%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e93.13%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e90.17%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e82.10%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003e\u003cb\u003eRMSProp\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGlioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e99.57%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.43%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e100.00%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e99.07%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e100.00%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e99.53%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e99.07%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMeningioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e98.48%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.52%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e98.06%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e95.28%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e99.44%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e96.65%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e93.52%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePituitary\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e98.91%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.09%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e96.55%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e100.00%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e98.44%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e98.25%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e96.55%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003e\u003cb\u003eNadam\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGlioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e99.57%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.43%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e100.00%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e99.07%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e100.00%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e99.53%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e99.07%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMeningioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e97.83%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e2.17%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e97.06%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e93.40%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e99.15%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e95.19%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e90.83%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePituitary\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e98.26%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.74%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e95.21%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e99.29%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e97.81%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e97.20%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e94.56%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab21\" class=\"InternalRef\"\u003e21\u003c/span\u003e shows the comparison of different evaluation parameters on different optimizers of the MobileNetV1 model regarding each class. The model achieved the best results on RMSProp Optimizer, with an accuracy of 99.57%, 98.48%, and 98.91% for glioma, meningioma, and pituitary tumor classes, respectively, with MCRs of 0.43%, 1.52%, and 1.09% for each class, respectively. While the model achieved the lowest results on the SGD optimizer, achieving an accuracy of 91.30%, 87.17%, and 93.70% for glioma, meningioma, and pituitary tumor classes, respectively, with MCRs of 8.70%, 12.83%, and 6.30% for each class respectively.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab22\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 22\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eOverall performance of MobileNetV1 model on different optimizers (Continue).\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"8\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOptimizers\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAccuracy\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eMCR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003ePrecision\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSensitivity\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eSpecificity\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eF1-Score\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eIoU\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eAdam\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e98.70%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.30%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e97.98%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e97.33%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e99.01%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e97.60%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e95.36%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eSGD\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e90.72%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e9.28%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e83.99%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e83.90%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e93.01%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e83.79%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e73.16%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eRMSProp\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e98.99%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.01%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e98.20%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e98.12%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e99.29%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e98.14%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e96.38%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eNadam\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e98.55%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.45%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e97.42%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e97.25%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e98.99%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e97.31%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e94.82%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab22\" class=\"InternalRef\"\u003e22\u003c/span\u003e shows the overall performance of the MobileNetV1 model on different optimizers regarding accuracy, MCR, precision, sensitivity, specificity, F1-score, and IoU. On all optimizers, RMSProp Optimizer produced the best results, and the model achieved results such as accuracy of 98.99%, MCR of 1.01%, precision of 98.20%, sensitivity of 98.12%, specificity of 99.29%, F1-score of 98.14%, and IoU of 96.38%. While the SGD optimizer produced the worst results and the model achieved an accuracy of 90.72% with MCR of 9.28%, precision of 83.99%, sensitivity of 83.90%, specificity of 93.01%, F1-score of 83.79%, and IoU of 73.16%. While Adam and Nadam optimizers produced moderate results better than SGD and model achieved results on Adam as accuracy of 98.70%, MCR of 1.30%, precision of 97.98%, sensitivity of 97.33%, specificity of 99.01%, F1-score of 97.60%, and IoU of 95.36%. while MobileNetV1 achieved results on Nadam Optimizer as accuracy of 98.55%, MCR of 1.45%, precision of 97.42%, sensitivity of 97.25%, specificity of 98.99%, F1-score of 97.31%, and IoU of 94.82%.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab23\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 23\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eOverall performance of MobileNetV1 model on different optimizers.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"9\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOptimizers\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNPV\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eFNR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eFPR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eFDR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eFOR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eMCC\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eBA\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c9\"\u003e \u003cp\u003eGM\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eAdam\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e99.13%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e2.67%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.99%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e2.02%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.87%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e96.42%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e98.17%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e98.14%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eSGD\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e93.11%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e16.10%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e6.99%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e16.01%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e6.89%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e77.10%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e88.45%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e88.09%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eRMSProp\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e99.26%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.88%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.71%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.80%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.74%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e97.40%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e98.70%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e98.69%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eNadam\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e98.97%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e2.75%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.01%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e2.58%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1.03%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e96.24%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e98.12%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e98.10%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab23\" class=\"InternalRef\"\u003e23\u003c/span\u003e shows the overall performance of the MobileNetV1 model regarding NPV, FNR, FDR, FOR, MCC, BA, and GM on different optimizers. As we see above, RMSProp Optimizer produced better results, Adam and Nadam produced moderate results, and SGD Optimizer produced the worst results. Regarding Table \u003cspan refid=\"Tab21\" class=\"InternalRef\"\u003e21\u003c/span\u003e performance parameters, the model achieved values on Adam Optimizer as NPV of 99.13%, FNR of 2.67%, FPR of 0.99%, FDR of 2.02%, FOR of 0.87%, MCC of 96.42%, BA of 98.17%, and GM of 98.14%. While on the SGD optimizer, the model achieved less results, as NPV of 93.11%, FNR of 16.10%, FPR of 6.99%, FDR of 16.01%, FOR of 6.89%, a MCC of 77.10%, a BA of 88.45%, and a GM of 88.09%. Further, the RMSProp model achieved a NPV of 99.26%, a FNR of 1.88%, FPR of 0.71%, FDR of 1.80%, FOR of 0.74%, MCC of 97.40%, BA and GM of 98.70%, and 98.69%, respectively. While on Nadam Optimizer, the model achieved results, as NPV of 98.97%, FNR of 2.75%, FPR of 1.01%, FDR of 2.58%, FOR of 1.03%, MCC of 96.24%, BA and GM of 98.12% and 98.10%, respectively.\u003c/p\u003e \u003cp\u003eBelow Table \u003cspan refid=\"Tab24\" class=\"InternalRef\"\u003e24\u003c/span\u003e shows the performance evaluation of the DenseNet121 model regarding different performance evaluation parameters, including accuracy, MCR, precision, sensitivity, specificity, F1-score, and IoU of each class on different optimizers, while Table \u003cspan refid=\"Tab25\" class=\"InternalRef\"\u003e25\u003c/span\u003e shows the overall performance of the DenseNet121 model regarding accuracy, MCR, precision, sensitivity, specificity, F1-score, and IoU, and Table \u003cspan refid=\"Tab26\" class=\"InternalRef\"\u003e26\u003c/span\u003e shows the overall performance of the DenseNet121 model regarding NPV, FNR, FPR, FDR, FOR, MCC, BA, and GM on different optimizers.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab24\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 24\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003ePerformance of DenseNet121 model regarding each class on different optimizers.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"9\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOptimizers\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eImage Class\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAccuracy\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eMCR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003ePrecision\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eSensitivity\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eSpecificity\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eF1-Score\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c9\"\u003e \u003cp\u003eIoU\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003e\u003cb\u003eAdam\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGlioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e99.35%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.65%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e100.00%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e98.60%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e100.00%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e99.29%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e98.60%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMeningioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e98.70%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.30%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e98.08%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e96.23%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e99.44%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e97.14%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e94.44%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePituitary\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e98.91%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.09%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e96.55%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e100.00%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e98.44%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e98.25%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e96.55%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003e\u003cb\u003eSGD\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGlioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e89.13%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e10.87%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e91.84%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e84.11%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e93.50%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e87.80%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e78.26%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMeningioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e84.57%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e15.43%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e69.23%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e59.43%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e92.09%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e63.96%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e47.01%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePituitary\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e90.65%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e9.35%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e78.03%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e96.43%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e88.13%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e86.26%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e75.84%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003e\u003cb\u003eRMSProp\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGlioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e99.13%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.87%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e99.53%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e98.60%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e99.59%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e99.06%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e98.14%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMeningioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e97.83%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e2.17%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e96.15%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e94.34%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e98.87%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e95.24%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e90.91%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePituitary\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e98.70%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.30%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e96.53%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e99.29%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e98.44%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e97.89%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e95.86%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003e\u003cb\u003eNadam\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGlioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e99.57%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.43%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e99.53%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e99.53%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e99.59%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e99.53%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e99.07%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMeningioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e98.70%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.30%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e99.02%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e95.28%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e99.72%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e97.12%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e94.39%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePituitary\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e99.13%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.87%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e97.22%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e100.00%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e98.75%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e98.59%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e97.22%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab24\" class=\"InternalRef\"\u003e24\u003c/span\u003e shows the comparison of different evaluation parameters on different optimizers of the DenseNet121 model regarding each class. The model achieved the best results on Nadam Optimizer, with an accuracy of 99.57%, 98.70%, and 99.13% for glioma, meningioma, and pituitary tumor classes, respectively, with MCRs of 0.43%, 1.30%, and 0.87% for each class, respectively. While the model achieved the lowest results on the SGD optimizer, achieving accuracy of 89.13%, 84.57%, and 90.65% for glioma, meningioma, and pituitary tumor classes, respectively, with MCRs of 10.87%, 15.43%, and 9.35% for each class respectively.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab25\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 25\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eOverall performance of DenseNet121 model on different optimizers (Continue).\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"8\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOptimizers\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAccuracy\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eMCR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003ePrecision\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSensitivity\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eSpecificity\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eF1-Score\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eIoU\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eAdam\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e98.99%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.01%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e98.21%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e98.27%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e99.29%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e98.23%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e96.53%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eSGD\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e88.12%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e11.88%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e79.70%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e79.99%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e91.24%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e79.34%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e67.04%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eRMSProp\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e98.55%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.45%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e97.40%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e97.41%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e98.97%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e97.40%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e94.97%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eNadam\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e99.13%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.87%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e98.59%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e98.27%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e99.35%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e98.41%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e96.89%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab25\" class=\"InternalRef\"\u003e25\u003c/span\u003e shows the overall performance of the DenseNet121 model on different optimizers regarding accuracy, MCR, precision, sensitivity, specificity, F1-score, and IoU. On all optimizers, Nadam Optimizer produced the best results, and the model achieved results such as accuracy of 99.13%, MCR of 0.87%, precision of 98.59%, sensitivity of 98.27%, specificity of 99.35%, F1-score of 98.41%, and IoU of 96.89%. While the SGD optimizer produced the worst results and the model achieved an accuracy of 88.12% with MCR of 11.88%, precision of 79.70%, sensitivity of 79.99%, specificity of 91.24%, F1-score of 79.34%, and IoU of 67.04%. While Adam and RMSProp optimizers produced moderate results better than SGD and model achieved results on Adam as accuracy of 98.99%, MCR of 1.01%, precision of 98.21%, sensitivity of 98.27%, specificity of 99.29%, F1-score of 98.23%, and IoU of 96.53%. while DenseNet121 achieved results on RMSProp Optimizer as accuracy of 98.55%, MCR of 1.45%, precision of 97.40%, sensitivity of 97.41%, specificity of 98.97%, F1-score of 97.40%, and IoU of 94.97%.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab26\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 26\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eOverall performance of DenseNet121 model on different optimizers.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"9\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOptimizers\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNPV\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eFNR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eFPR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eFDR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eFOR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eMCC\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eBA\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c9\"\u003e \u003cp\u003eGM\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eAdam\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e99.22%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.73%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.71%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.79%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.78%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e97.54%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e98.78%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e98.78%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eSGD\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e91.24%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e20.01%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e8.76%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e20.30%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e8.76%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e71.55%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e85.61%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e84.95%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eRMSProp\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e98.93%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e2.59%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.03%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e2.60%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1.07%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e96.36%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e98.19%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e98.18%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eNadam\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e99.40%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.73%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.65%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.41%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.60%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e97.66%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e98.81%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e98.80%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab26\" class=\"InternalRef\"\u003e26\u003c/span\u003e shows the overall performance of the DenseNet121 model regarding NPV, FNR, FDR, FOR, MCC, BA, and GM on different optimizers. As we see above, Nadam Optimizer produced better results, Adam and RMSProp produced moderate results, and SGD Optimizer produced the worst results. Regarding Table \u003cspan refid=\"Tab24\" class=\"InternalRef\"\u003e24\u003c/span\u003e performance parameters, the model achieved values on Adam Optimizer as NPV of 99.22%, FNR of 1.73%, FPR of 0.71%, FDR of 1.79%, FOR of 0.78%, MCC of 97.54%, BA of 98.78%, and GM of 98.78%. While on the SGD optimizer, the model achieved less results, as NPV of 91.24%, FNR of 20.01%, FPR of 8.76%, FDR of 20.30%, FOR of 8.76%, a MCC of 71.55%, a BA of 85.61%, and a GM of 84.95%. Further, the RMSProp model achieved a NPV of 98.93%, a FNR of 2.59%, FPR of 1.03%, FDR of 2.60%, FOR of 1.07%, MCC of 96.36%, BA and GM of 98.19%, and 98.18%, respectively. While on Nadam Optimizer, the model achieved results, as NPV of 99.40%, FNR of 1.73%, FPR of 0.65%, FDR of 1.14%, FOR of 0.60%, MCC of 97.66%, BA and GM of 98.81% and 98.80%, respectively.\u003c/p\u003e \u003cp\u003eAdditionally, Table \u003cspan refid=\"Tab27\" class=\"InternalRef\"\u003e27\u003c/span\u003e shows the performance evaluation of the EfficientNetB0 model regarding different performance evaluation parameters, including accuracy, MCR, precision, sensitivity, specificity, F1-score, and IoU of each class on different optimizers, while Table \u003cspan refid=\"Tab28\" class=\"InternalRef\"\u003e28\u003c/span\u003e shows the overall performance of the EfficientNetB0 model regarding accuracy, MCR, precision, sensitivity, specificity, F1-score, and IoU, and Table \u003cspan refid=\"Tab29\" class=\"InternalRef\"\u003e29\u003c/span\u003e shows the overall performance of the EfficientNetB0 model regarding NPV, FNR, FPR, FDR, FOR, MCC, BA, and GM on different optimizers.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab27\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 27\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003ePerformance of EfficientNetB0 model regarding each class on different optimizers.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"9\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOptimizers\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eImage Class\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAccuracy\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eMCR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003ePrecision\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eSensitivity\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eSpecificity\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eF1-Score\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c9\"\u003e \u003cp\u003eIoU\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003e\u003cb\u003eAdam\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGlioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e100.00%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.00%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e100.00%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e100.00%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e100.00%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e100.00%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e100.00%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMeningioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e98.70%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.30%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e100.00%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e94.34%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e100.00%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e97.09%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e94.34%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePituitary\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e98.70%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.30%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e95.89%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e100.00%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e98.13%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e97.90%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e95.89%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003e\u003cb\u003eSGD\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGlioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e58.48%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e41.52%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e52.96%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e96.26%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e25.61%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e68.33%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e51.89%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMeningioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e76.52%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e23.48%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e37.50%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e2.83%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e98.59%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e5.26%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e2.70%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePituitary\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e73.26%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e26.74%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e63.49%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e28.57%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e92.81%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e39.41%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e24.54%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003e\u003cb\u003eRMSProp\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGlioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e99.35%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.65%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e99.53%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e99.07%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e99.59%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e99.30%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e98.60%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMeningioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e98.26%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.74%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e100.00%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e92.45%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e100.00%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e96.08%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e92.45%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePituitary\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e98.04%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.96%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e93.96%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e100.00%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e97.19%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e96.89%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e93.96%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003e\u003cb\u003eNadam\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGlioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e98.91%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.09%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e98.16%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e99.53%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e98.37%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e98.84%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e97.71%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMeningioma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e97.61%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e2.39%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e100.00%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e89.62%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e100.00%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e94.53%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e89.62%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePituitary\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e98.26%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.74%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e94.59%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e100.00%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e97.50%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e97.22%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e94.59%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab27\" class=\"InternalRef\"\u003e27\u003c/span\u003e shows the comparison of different evaluation parameters on different optimizers of the EfficientNetB0 model regarding each class. The model achieved the best results on Adam Optimizer, with an accuracy of 100%, 98.70%, and 98.70% for glioma, meningioma, and pituitary tumor classes, respectively, with MCRs of 0.0%, 1.30%, and 1.30% for each class, respectively. While the model achieved the lowest results on the SGD optimizer, achieving accuracy of 58.48%, 76.52%, and 73.26% for glioma, meningioma, and pituitary tumor classes, respectively, with MCRs of 41.52%, 23.48%, and 26.74% for each class respectively.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab28\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 28\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eOverall performance of EfficientNetB0 model on different optimizers (Continue).\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"8\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOptimizers\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAccuracy\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eMCR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003ePrecision\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSensitivity\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eSpecificity\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eF1-Score\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eIoU\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eAdam\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e99.13%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.87%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e98.63%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e98.11%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e99.38%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e98.33%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e96.74%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eSGD\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e69.42%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e30.58%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e51.32%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e42.55%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e72.34%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e37.67%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e26.38%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eRMSProp\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e98.55%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.45%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e97.83%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e97.17%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e98.93%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e97.42%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e95.01%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eNadam\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e98.26%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.74%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e97.58%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e96.39%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e98.62%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e96.86%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e93.97%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab28\" class=\"InternalRef\"\u003e28\u003c/span\u003e shows the overall performance of the EfficientNetB0 model on different optimizers regarding accuracy, MCR, precision, sensitivity, specificity, F1-score, and IoU. On all optimizers, Adam Optimizer produced the best results, and the model achieved results such as accuracy of 99.13%, MCR of 0.87%, precision of 98.63%, sensitivity of 98.11%, specificity of 99.38%, F1-score of 98.33%, and IoU of 96.74%. While the SGD optimizer produced the worst results and the model achieved an accuracy of 69.42% with MCR of 30.58%, precision of 51.32%, sensitivity of 42.55%, specificity of 72.34%, F1-score of 37.67%, and IoU of 26.38%. While Nadam and RMSProp optimizers produced moderate results better than SGD and model achieved results on RMSProp as accuracy of 98.55%, MCR of 1.45%, precision of 97.83%, sensitivity of 97.17%, specificity of 98.93%, F1-score of 97.42%, and IoU of 95.01%. while EfficientNetB0 achieved results on Nadam Optimizer as accuracy of 98.26%, MCR of 1.74%, precision of 97.58%, sensitivity of 96.39%, specificity of 98.62%, F1-score of 98.86%, and IoU of 93.97%.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab29\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 29\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eOverall performance of EfficientNetB0 model on different optimizers.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"9\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOptimizers\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNPV\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eFNR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eFPR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eFDR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eFOR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eMCC\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eBA\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c9\"\u003e \u003cp\u003eGM\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eAdam\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e99.44%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.89%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.63%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.37%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.56%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e97.54%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e98.74%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e98.73%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eSGD\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e80.25%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e57.45%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e27.66%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e48.68%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e19.75%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e20.39%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e57.45%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e39.28%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eRMSProp\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e98.99%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e2.83%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.07%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e2.17%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1.01%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e96.16%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e98.05%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e98.02%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eNadam\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e98.86%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e3.61%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.38%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e2.42%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1.14%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e95.17%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e97.50%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e97.45%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab29\" class=\"InternalRef\"\u003e29\u003c/span\u003e shows the overall performance of the EfficientNetB0 model regarding NPV, FNR, FDR, FOR, MCC, BA, and GM on different optimizers. As we see above, Adam Optimizer produced better results, Nadam and RMSProp produced moderate results, and SGD Optimizer produced the worst results. Regarding Table \u003cspan refid=\"Tab27\" class=\"InternalRef\"\u003e27\u003c/span\u003e performance parameters, the model achieved values on Adam Optimizer as NPV of 99.44%, FNR of 1.89%, FPR of 0.63%, FDR of 1.37%, FOR of 0.56%, MCC of 97.54%, BA of 98.74%, and GM of 98.73%. While on the SGD optimizer, the model achieved less results, as NPV of 80.25%, FNR of 57.45%, FPR of 27.66%, FDR of 48.68%, FOR of 19.75%, a MCC of 20.39%, a BA of 57.45%, and a GM of 39.28%. Further, the RMSProp model achieved a NPV of 98.99%, a FNR of 2.83%, FPR of 1.07%, FDR of 2.17%, FOR of 1.01%, MCC of 96.16%, BA and GM of 98.05%, and 98.02%, respectively. While on Nadam Optimizer, the model achieved results, as NPV of 98.86%, FNR of 3.61%, FPR of 1.38%, FDR of 2.42%, FOR of 1.14%, MCC of 95.17%, BA and GM of 97.50% and 97.45%, respectively. Further, Tables \u003cspan refid=\"Tab30\" class=\"InternalRef\"\u003e30\u003c/span\u003e and \u003cspan refid=\"Tab31\" class=\"InternalRef\"\u003e31\u003c/span\u003e show the overall performance of the models in terms of achieving the best results with respect to optimizers regarding different statistical performance evaluation parameters that were already discussed above.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab30\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 30\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eOverall Performance of the Models Regarding Best Results Producing Optimizers (Continue).\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"9\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eModel Architecture\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eOptimizers\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAccuracy\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eMCR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003ePrecision\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eSensitivity\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eSpecificity\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eF1-Score\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c9\"\u003e \u003cp\u003eIoU\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eAlexNet\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAdam\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e95.94%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e4.06%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e94.38%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e92.29%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e96.55%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e93.19%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e87.47%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eVGG16\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNadam\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e98.41%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.59%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e97.01%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e97.56%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e98.89%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e97.27%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e94.72%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eVGG19\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003eAdam\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003e99.13%\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e\u003cb\u003e0.87%\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e98.58%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u003cb\u003e98.43%\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e99.33%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e\u003cb\u003e98.50%\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e\u003cb\u003e97.06%\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eResNet50\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAdam\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e98.41%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.59%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e97.33%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e97.02%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e98.82%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e97.14%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e94.49%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eXception\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eRMSProp\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e98.99%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.01%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e98.20%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e98.03%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e99.29%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e98.10%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e96.30%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eInception V3\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eRMSProp\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e98.84%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.16%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e97.95%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e97.88%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e99.17%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e97.91%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e95.93%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eDenseNet121\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003eNadam\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003e99.13%\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e\u003cb\u003e0.87%\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e98.59%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e98.27%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e99.35%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e98.41%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e96.89%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eMobileNetV1\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eRMSProp\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e98.99%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.01%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e98.20%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e98.12%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e99.29%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e98.14%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e96.38%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eEfficientNetB0\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003eAdam\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003e99.13%\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e\u003cb\u003e0.87%\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e98.63%\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e98.11%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e\u003cb\u003e99.38%\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e98.33%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e96.74%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab31\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 31\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eOverall Performance of the Models Regarding Best Results Producing Optimizers.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"10\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c10\" colnum=\"10\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eModel Architecture\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eOptimizers\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNPV\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eFNR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eFPR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eFDR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eFOR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eMCC\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c9\"\u003e \u003cp\u003eBA\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c10\"\u003e \u003cp\u003eGM\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eAlexNet\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAdam\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e97.16%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e7.71%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e3.45%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e5.62%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e2.84%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e89.20%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e94.42%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e94.29%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eVGG16\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNadam\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e98.74%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e2.44%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.11%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e2.99%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e1.26%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e96.38%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e98.23%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e98.23%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eVGG19\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003eAdam\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e99.35%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e\u003cb\u003e1.57%\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.67%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1.42%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.65%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e\u003cb\u003e97.78%\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e\u003cb\u003e98.88%\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e\u003cb\u003e98.88%\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eResNet50\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAdam\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e98.85%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e2.98%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.18%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e2.67%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e1.15%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e95.87%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e97.92%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e97.90%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eXception\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eRMSProp\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e99.29%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.97%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.71%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1.80%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.71%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e97.33%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e98.66%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e98.65%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eInception V3\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eRMSProp\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e99.16%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e2.12%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.83%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e2.05%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.84%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e97.04%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e98.52%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e98.52%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eDenseNet121\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003eNadam\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e99.40%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.73%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.65%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1.41%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.60%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e97.66%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e98.81%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e98.80%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eMobileNetV1\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eRMSProp\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e99.26%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.88%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.71%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1.80%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.74%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e97.40%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e98.70%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e98.69%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eEfficientNetB0\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003eAdam\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003e99.44%\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.89%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e0.63%\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u003cb\u003e1.37%\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e\u003cb\u003e0.56%\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e97.54%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e98.74%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e98.73%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eIn the above Tables \u003cspan refid=\"Tab30\" class=\"InternalRef\"\u003e30\u003c/span\u003e and \u003cspan refid=\"Tab31\" class=\"InternalRef\"\u003e31\u003c/span\u003e, the overall results of different models regarding different optimizers are shown with respect to the different performance evaluation parameters shown. All the details of the overall performance of the models regarding different optimizers are already discussed in the above tables, while the AlexNet, VGG19, ResNet50, and EfficientNetB0 models produced the overall best results on Adam optimizer by achieving an overall accuracy of 95.94%, 99.13%, 98.41%, and 99.13% with MCRs of 4.06%, 0.87%, 1.59%, and 0.87%, respectively. While VGG16 and DenseNet121 produced the best results on Nadam Optimizer by achieving accuracy of 98.41% and 99.13% with MCRs of 1.59% and 0.87%, respectively, Further, Xception, InceptionV3, and MobileNetV1 produced better results on the RMSProp optimizer, achieving an overall accuracy of 98.99%, 98.84%, and 98.99% with MCRs of 1.01%, 1.16%, and 1.88%, respectively. None of the pre-trained models produced better results on the SGD optimizer, and among all the models regarding optimizers, the Adam optimizer works best, and four different models produced better results on it. While among all the different models, three produced almost the same results on different optimizers, there was a slight difference in their performance evaluation parameters. VGG19 on Adam Optimizer, DenseNet121 on Nadam Optimizer, and EfficientNetB0 on Adam Optimizer produce the same accuracy and MCR, achieving an accuracy of 99.13% and an MCR of 0.87%, respectively. While the EfficientNetB0 model produced better precision and specificity than other models at 98.63% and 99.38%, respectively, VGG19 produced better sensitivity, F1-score, and IoU than other models at 98.43%, 98.50%, and 97.06%, respectively. While EfficientNetB0 produced better results on NPV at 99.44%, which is higher than other models achieved values, and better results on FPR, FDR, and FOR as 0.63%, 1.37%, and 0.56%, respectively, which is less than other models achieved values. Further, VGG19 produced better results on FNR, MCC, BA, and GM, with a 1.57% FNR, which is less than other models achieved, and 97.78%, 98.88%, and 98.88%, respectively, which are higher than other models\u0026rsquo; values. Overall, we can predict that the VGG19 and EfficientNetB0 models achieved better results on the Adam optimizer. Further, Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e shows the overall performance of different pre-trained models with the best result-producing optimizer regarding accuracy and MCR. Further, the confusion matrix (CM) of the best results-producing optimizers on deep learning models is shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e4\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eFigure \u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e4\u003c/span\u003e shows the confusion matrices of all the pre-trained models with the best results, producing optimizers. The confusion matrix provides detailed counts of correct and incorrect predictions across all classes, which give insights into the model overall performance. Figure\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e shows the receiver operating character (ROC) curves of the VGG19 and EfficientNetB0 models, which achieved the overall best performances among different pre-trained models on different optimizers, and this curve represents the trade-off between true and false positive rates across all possible threshold classifications by summarizing performance in the area under curve (AUC) metric.\u003c/p\u003e\u003cp\u003eFigure 5 shows the results of the VGG19 and EfficientNetB0 models on Adam optimizers in the form of ROC curves, respectively, while the VGG19 model produced AUC for all three classes as 1.00, 0.99, and 1.00 for pituitary, meningioma, and glioma classes, respectively, while the EfficientNetB0 model produced AUC for all three classes as 1.00, which shows the superiority of these two models on Adam optimizers over other models. Further, below figure 6 is the testing result of the best-producing deep learning models with their testing accuracy, actual and predicted class labels against three brain tumor classes, respectively.\u003c/p\u003e \u003cp\u003eSo above Fig.\u0026nbsp;6 shows the testing results of glioma, meningioma, and pituitary tumor with a testing accuracy of 97.61% with actual and predicted labels that show the ability of deep learning-based approaches to diagnose diseases accurately in humans to cure their problems and save the human\u0026rsquo;s life.\u003c/p\u003e \u003cp\u003eFrom above all the models, EfficientNetB0 and VGG19 with Adam Optimizer perform well just because of the combined benefits of using both momentum, which accelerates the SGD by focusing on directionally consisting updates, and RMSProp, which adjusts the learning rate based on recenet gradient. DenseNet121 with Nadam Optimizor performs well just due to the incorporation of Nesterov momentum, which looks ahead by computing the gradient at the predicted future position. Adam optimizer overcomes issues like slow convergence or vanishing gradient problems, which allow the VGG19 model to perform well in classification tasks as well as for EfficientNetB0, while Nadam\u0026rsquo;s predictive updates fit well with densely connected layers of DenseNet121 that benefit from more accurate gradient updates, which allow the model to reuse features and parameters that enhance the performance on classification.\u003c/p\u003e \u003cp\u003eFurther, the below Table \u003cspan refid=\"Tab32\" class=\"InternalRef\"\u003e32\u003c/span\u003e shows the comparison of the proposed best result achievable model with other state-of-the-art approaches grading different statistical performance evaluation parameters.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab32\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 32\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eProposed model result comparison with other state-of-the-art approaches.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"9\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRef\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTechniques\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eDataset\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAccuracy\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eMCR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003ePrecision\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eSensitivity\u003c/p\u003e \u003cp\u003e/Recall\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eSpecificity\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c9\"\u003e \u003cp\u003eF1-Score\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e[\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e23-CNN layers with VGG16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eFigshare\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e97.8%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e2.2%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e96.5%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e96.4%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eNill\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e96.4%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e[\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePDCNN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eFigshare\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e97.60%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e2.40%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e97%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e97%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eNill\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e97%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e[\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eEnsemble ViTs\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eFigshare\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e98.70%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.30%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eNill\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e97.78%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e99.42%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003eNill\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e[\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDenseNet41-cornerNet\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eFigshare\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e98.8%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.20%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eNill\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eNill\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eNill\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003eNill\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e[\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMultimodal-SVM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eFigshare\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e98.92%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.08%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eNill\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e98.82%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e99.02%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003eNill\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e[\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eVGGNet\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eFigshare\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e98.93%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.07%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e99.11%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e98.68%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e99.13%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003eNill\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e[\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDSRUF\u0026thinsp;+\u0026thinsp;HOG\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eFigshare\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e90.27%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e9.73%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eNill\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eNill\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eNill\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003eNill\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e[\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eResNet50 - Adadelta\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eFigshare\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e99.02%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.98%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eNill\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eNill\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eNill\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003eNill\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003eKhushi et al.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eVGG19 - Adam\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eFigshare\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e99.13%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.87%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e98.58%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e98.43%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e99.33%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e98.50%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDenseNet121 - Nadam\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eFigshare\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e99.13%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.87%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e98.59%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e98.27%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e99.35%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e98.41%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eEfficientNetB0 - Adam\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eFigshare\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e99.13%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.87%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e98.63%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e98.11%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e99.38%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e98.33%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab32\" class=\"InternalRef\"\u003e32\u003c/span\u003e shows the comparison of our best achievable results models regarding optimizers with other state-of-the art approaches. From above all models on four different optimizers, VGG19 with Adam, DenseNet121 with Nadam, and Effi-cientNetB0 with Adam optimizer achieved the best results in terms of accuracy and miss classification rate that all other state-of-the-art approaches by achieving 99.13% accuracy with 0.87% MCR. VGG19-Adam achieved the best sensitivity or recall value of 98.43% and the F1-Score value of 98.50%, which is better than our other models as well as with other approaches, including [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e], [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e], [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e], and all other Nill in terms of sensitivity and better than other approaches including [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e], [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e], and all other Nill in terms of F1-score, respectively. Further, EfficientNetB0-Adam achieved the best result in terms of precision and specificity by achieving values of 98.63% and 99.38%, respectively, which is better than [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e], [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e], and all other Nill value approaches in terms of precision and better than [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e], [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e], and all other Nill approaches in terms of specificity, respectively. Further, all three models achieved the best accuracy and low MCR with other state-of-the-art approaches mentioned in Table\u0026nbsp;\u003cspan refid=\"Tab32\" class=\"InternalRef\"\u003e32\u003c/span\u003e, which shows the superiority of our proposed research work. The above three models produced the same results regarding accuracy and MCR, but the other performance evaluation parameters results are changed, and every model produced different results regarding each class, but unfortunately the average results of these three models are the same only on accuracy and MCR. The reason is that each model is highly robust and well optimized, and their ability to learn and generalize from the data is comparable across different architectures when applied to the same dataset, and different optimizers help models reach a similar level of optimization, which leads toward similar performance, particularly on well-defined datasets. Overall, each model performed differently against each class.\u003c/p\u003e"},{"header":"Conclusions","content":"\u003cp\u003eIn recent decades, brain tumors have become one of the deadliest diseases, with survival rates decreasing over time. Diagnosing brain tumors has become increasingly challenging for paramedical staff. Early detection is crucial for improving survival rates. To address this, we utilized various pre-trained deep learning models with different optimizers to facilitate the early diagnosis of brain tumors. We evaluated the performance of deep learning models, including AlexNet, VGG16, VGG19, ResNet50, Xception, InceptionV3, DenseNet121, MobileNetV1, and EfficientNetB0, using optimizers such as Adam, SGD, RMSProp, and Nadam, on the publicly available Figshare brain tumor dataset. Among the models tested, VGG19 and EfficientNetB0 performed exceptionally well with the Adam optimizer, achieving an accuracy of 99.13% and a misclassification rate (MCR) of 0.87%. Both models also demonstrated strong performance in other statistical evaluation parameters and ROC curve analysis. EfficientNetB0 outperformed the others, achieving an area under the curve (AUC) of 100% for each tumor class and testing an accuracy of 97.61% on random images.\u003c/p\u003e"},{"header":"Limitations and future work","content":"\u003cp\u003eThe main purpose of this research was to evaluate various pre-trained deep learning models with different optimizers for the accurate detection and diagnosis of brain tumors using the publicly available Figshare dataset. Additionally, we plan to explore more deep learning-based models on this and other brain tumor datasets, as well as for diagnosing other diseases in the human body, with the aim of improving human survival.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eFunding: \u003c/strong\u003eThis research work have no funding.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData Availability Statement: \u003c/strong\u003eThe datasets that have been used in this research work have been taken from an open-source library, Kaggle: https://www.kaggle.com/datasets/ashkhagan/figshare-brain-tumor-dataset \u0026amp; Simulation files used during the current study are available from the corresponding author upon reasonable request.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConflicts of Interest: \u003c/strong\u003eThe authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor Contribution: \u003c/strong\u003eConceptualization, Writing original draft, Experimentation: H.M.T.K; Supervision: T.M; Validation, Proof reading: I.N and T.M.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eN. Varuna Shree and T. N. R. Kumar, \u0026ldquo;Identification and classification of brain tumor MRI images with feature extraction using DWT and probabilistic neural network,\u0026rdquo; \u003cem\u003eBrain Informatics\u003c/em\u003e, vol. 5, no. 1, pp. 23\u0026ndash;30, 2018, doi: 10.1007/s40708-017-0075-5.\u003c/li\u003e\n\u003cli\u003eA. Bunevicius, K. Schregel, R. Sinkus, A. Golby, and S. Patz, \u0026ldquo;REVIEW: MR elastography of brain tumors,\u0026rdquo; \u003cem\u003eNeuroImage Clin.\u003c/em\u003e, vol. 25, no. October 2019, 2020, doi: 10.1016/j.nicl.2019.102109.\u003c/li\u003e\n\u003cli\u003eM. Santos and N. P. Rocha, \u0026ldquo;Medical Imaging Repository Contributions for Radiation Protection Key Performance Indicators,\u0026rdquo; \u003cem\u003eProcedia Comput. Sci.\u003c/em\u003e, vol. 196, pp. 590\u0026ndash;597, 2021, doi: 10.1016/j.procs.2021.12.053.\u003c/li\u003e\n\u003cli\u003eI. Naseer, S. Akram, T. Masood, A. Jaffar, M. A. Khan, and A. Mosavi, \u0026ldquo;Performance Analysis of State-of-the-Art CNN Architectures for LUNA16,\u0026rdquo; \u003cem\u003eSensors\u003c/em\u003e, vol. 22, no. 12, 2022, doi: 10.3390/s22124426.\u003c/li\u003e\n\u003cli\u003eH. M. T. Khushi, T. Masood, A. Jaffar, M. Rashid, and S. Akram, \u0026ldquo;Improved Multiclass Brain Tumor Detection via Customized Pretrained EfficientNetB7 Model,\u0026rdquo; \u003cem\u003eIEEE Access\u003c/em\u003e, vol. 11, no. October, pp. 117210\u0026ndash;117230, 2023, doi: 10.1109/ACCESS.2023.3325883.\u003c/li\u003e\n\u003cli\u003eH. M. T. Khushi, T. Masood, A. Jaffar, S. Akram, and S. M. Bhatti, \u0026ldquo;Performance analysis of state-of-the-art CNN architectures for brain tumour detection,\u0026rdquo; \u003cem\u003eInt. J. Imaging Syst. Technol.\u003c/em\u003e, vol. 34, no. 1, pp. 1\u0026ndash;15, 2024, doi: 10.1002/ima.22949.\u003c/li\u003e\n\u003cli\u003eM. A. Talukder \u003cem\u003eet al.\u003c/em\u003e, \u0026ldquo;An efficient deep learning model to categorize brain tumor using reconstruction and fine-tuning,\u0026rdquo; \u003cem\u003eExpert Syst. Appl.\u003c/em\u003e, vol. 230, no. August 2022, p. 120534, 2023, doi: 10.1016/j.eswa.2023.120534.\u003c/li\u003e\n\u003cli\u003eM. S. I. Khan \u003cem\u003eet al.\u003c/em\u003e, \u0026ldquo;Accurate brain tumor detection using deep convolutional neural network,\u0026rdquo; \u003cem\u003eComput. Struct. Biotechnol. J.\u003c/em\u003e, vol. 20, pp. 4733\u0026ndash;4745, 2022, doi: 10.1016/j.csbj.2022.08.039.\u003c/li\u003e\n\u003cli\u003eT. Rahman and M. S. Islam, \u0026ldquo;MRI brain tumor detection and classification using parallel deep convolutional neural networks,\u0026rdquo; \u003cem\u003eMeas. Sensors\u003c/em\u003e, vol. 26, no. February, p. 100694, 2023, doi: 10.1016/j.measen.2023.100694.\u003c/li\u003e\n\u003cli\u003eN. Ullah \u003cem\u003eet al.\u003c/em\u003e, \u0026ldquo;An Effective Approach to Detect and Identify Brain Tumors Using Transfer Learning,\u0026rdquo; \u003cem\u003eAppl. Sci.\u003c/em\u003e, vol. 12, no. 11, 2022, doi: 10.3390/app12115645.\u003c/li\u003e\n\u003cli\u003eS. Tummala, S. Kadry, S. A. C. Bukhari, and H. T. Rauf, \u0026ldquo;Classification of Brain Tumor from Magnetic Resonance Imaging Using Vision Transformers Ensembling,\u0026rdquo; \u003cem\u003eCurr. Oncol. 2022, Vol. 29, Pages 7498-7511\u003c/em\u003e, vol. 29, no. 10, pp. 7498\u0026ndash;7511, Oct. 2022, doi: 10.3390/CURRONCOL29100590.\u003c/li\u003e\n\u003cli\u003eM. Nawaz \u003cem\u003eet al.\u003c/em\u003e, \u0026ldquo;[94] Analysis of Brain MRI Images Using Improved 2021.pdf,\u0026rdquo; pp. 1\u0026ndash;18, 2021.\u003c/li\u003e\n\u003cli\u003eS. Maqsood, R. Dama\u0026scaron;evičius, and R. Maskeliūnas, \u0026ldquo;Multi-Modal Brain Tumor Detection Using Deep Neural Network and Multiclass SVM,\u0026rdquo; \u003cem\u003eMed.\u003c/em\u003e, vol. 58, no. 8, 2022, doi: 10.3390/medicina58081090.\u003c/li\u003e\n\u003cli\u003eP. P. Malla, S. Sahu, and A. I. Alutaibi, \u0026ldquo;Classification of Tumor in Brain MR Images Using Deep Convolutional Neural Network and Global Average Pooling,\u0026rdquo; \u003cem\u003eProcesses\u003c/em\u003e, vol. 11, no. 3, p. 679, 2023, doi: 10.3390/pr11030679.\u003c/li\u003e\n\u003cli\u003eW. Ayadi, I. Charfi, W. Elhamzi, and M. Atri, \u0026ldquo;Brain tumor classification based on hybrid approach,\u0026rdquo; \u003cem\u003eVis. Comput.\u003c/em\u003e, vol. 38, no. 1, pp. 107\u0026ndash;117, 2022, doi: 10.1007/s00371-020-02005-1.\u003c/li\u003e\n\u003cli\u003e\u0026Ouml;. Polat and C. G\u0026uuml;ngen, \u0026ldquo;Classification of brain tumors from MR images using deep transfer learning,\u0026rdquo; \u003cem\u003eJ. Supercomput.\u003c/em\u003e, vol. 77, no. 7, pp. 7236\u0026ndash;7252, 2021, doi: 10.1007/s11227-020-03572-9.\u003c/li\u003e\n\u003cli\u003eM. Agarwal, G. Rani, A. Kumar, P. K. K, R. Manikandan, and A. H. Gandomi, \u0026ldquo;Deep learning for enhanced brain Tumor Detection and classification,\u0026rdquo; \u003cem\u003eResults Eng.\u003c/em\u003e, vol. 22, no. December 2023, p. 102117, 2024, doi: 10.1016/j.rineng.2024.102117.\u003c/li\u003e\n\u003cli\u003eA. Bhuvaneswari Ramakrishnan, M. Sridevi, S. K. Vasudevan, R. Manikandan, and A. H. Gandomi, \u0026ldquo;Optimizing brain tumor classification with hybrid CNN architecture: Balancing accuracy and efficiency through oneAPI optimization,\u0026rdquo; \u003cem\u003eInformatics Med. Unlocked\u003c/em\u003e, vol. 44, p. 101436, Jan. 2024, doi: 10.1016/J.IMU.2023.101436.\u003c/li\u003e\n\u003cli\u003eR. Sekaran, A. K. Munnangi, M. Ramachandran, and A. H. Gandomi, \u0026ldquo;3D brain slice classification and feature extraction using Deformable Hierarchical Heuristic Model,\u0026rdquo; \u003cem\u003eComput. Biol. Med.\u003c/em\u003e, vol. 149, p. 105990, Oct. 2022, doi: 10.1016/J.COMPBIOMED.2022.105990.\u003c/li\u003e\n\u003cli\u003eA. Kumar, M. Ramachandran, A. H. Gandomi, R. Patan, S. Lukasik, and R. K. Soundarapandian, \u0026ldquo;A deep neural network based classifier for brain tumor diagnosis,\u0026rdquo; \u003cem\u003eAppl. Soft Comput.\u003c/em\u003e, vol. 82, p. 105528, Sep. 2019, doi: 10.1016/J.ASOC.2019.105528.\u003c/li\u003e\n\u003cli\u003e\u0026ldquo;Figshare Brain Tumor Dataset.\u0026rdquo; Accessed: Jun. 15, 2024. [Online]. Available: https://www.kaggle.com/datasets/ashkhagan/figshare-brain-tumor-dataset\u003c/li\u003e\n\u003cli\u003eK. R. Avery \u003cem\u003eet al.\u003c/em\u003e, \u0026ldquo;ImageNet classification with deep convolutional neural networks,\u0026rdquo; \u003cem\u003eCommun. ACM\u003c/em\u003e, vol. 7, no. 3, pp. 84\u0026ndash;90, Apr. 2017, doi: 10.1145/3065386.\u003c/li\u003e\n\u003cli\u003eK. Simonyan and A. Zisserman, \u0026ldquo;Very deep convolutional networks for large-scale image recognition,\u0026rdquo; \u003cem\u003e3rd Int. Conf. Learn. Represent. ICLR 2015 - Conf. Track Proc.\u003c/em\u003e, pp. 1\u0026ndash;14, 2015.\u003c/li\u003e\n\u003cli\u003eK. He, X. Zhang, S. Ren, and J. Sun, \u0026ldquo;Deep Residual Learning for Image Recognition\u0026rdquo;, Accessed: Jun. 16, 2024. [Online]. Available: http://image-net.org/challenges/LSVRC/2015/\u003c/li\u003e\n\u003cli\u003eF. Chollet, \u0026ldquo;Xception: Deep Learning with Depthwise Separable Convolutions\u0026rdquo;.\u003c/li\u003e\n\u003cli\u003eC. Szegedy, V. Vanhoucke, S. Ioffe, J. Shlens, and Z. Wojna, \u0026ldquo;Rethinking the Inception Architecture for Computer Vision,\u0026rdquo; \u003cem\u003eProc. IEEE Comput. Soc. Conf. Comput. Vis. Pattern Recognit.\u003c/em\u003e, vol. 2016-Decem, pp. 2818\u0026ndash;2826, 2016, doi: 10.1109/CVPR.2016.308.\u003c/li\u003e\n\u003cli\u003eA. G. Howard \u003cem\u003eet al.\u003c/em\u003e, \u0026ldquo;MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications,\u0026rdquo; Apr. 2017, Accessed: Dec. 25, 2023. [Online]. Available: https://arxiv.org/abs/1704.04861v1\u003c/li\u003e\n\u003cli\u003eG. Huang, Z. Liu, L. Van Der Maaten, and K. Q. Weinberger, \u0026ldquo;Densely Connected Convolutional Networks\u0026rdquo;, Accessed: Jun. 16, 2024. [Online]. Available: https://github.com/liuzhuang13/DenseNet.\u003c/li\u003e\n\u003cli\u003eM. Tan and Q. V. Le, \u0026ldquo;EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks,\u0026rdquo; \u003cem\u003eInt. Conf. Mach. Learn.\u003c/em\u003e, 2019.\u003c/li\u003e\n\u003cli\u003eK. Chandra, A. Xie, J. Ragan-Kelley, and E. MEIJER, \u0026ldquo;Gradient Descent: The Ultimate Optimizer,\u0026rdquo; \u003cem\u003eAdv. Neural Inf. Process. Syst.\u003c/em\u003e, vol. 35, pp. 8214\u0026ndash;8225, Dec. 2022.\u003c/li\u003e\n\u003cli\u003eI. Bello, B. Zoph, V. Vasudevan, and Q. V. Le, \u0026ldquo;Neural Optimizer Search with Reinforcement Learning.\u0026rdquo; PMLR, pp. 459\u0026ndash;468, Jul. 17, 2017. Accessed: Jun. 20, 2024. [Online]. Available: https://proceedings.mlr.press/v70/bello17a.html\u003c/li\u003e\n\u003cli\u003eZ. Zhang, \u0026ldquo;Improved Adam Optimizer for Deep Neural Networks,\u0026rdquo; \u003cem\u003e2018 IEEE/ACM 26th Int. Symp. Qual. Serv. IWQoS 2018\u003c/em\u003e, Jan. 2019, doi: 10.1109/IWQOS.2018.8624183.\u003c/li\u003e\n\u003cli\u003eQ. Zhang \u003cem\u003eet al.\u003c/em\u003e, \u0026ldquo;Boosting Adversarial Attacks with Nadam Optimizer,\u0026rdquo; \u003cem\u003eElectron. 2023, Vol. 12, Page 1464\u003c/em\u003e, vol. 12, no. 6, p. 1464, Mar. 2023, doi: 10.3390/ELECTRONICS12061464.\u003c/li\u003e\n\u003cli\u003eH. M. Shahzad, S. M. Bhatti, A. Jaffar, M. Rashid, and S. Akram, \u0026ldquo;Multi-Modal CNN Features Fusion for Emotion Recognition: A Modified Xception Model,\u0026rdquo; \u003cem\u003eIEEE Access\u003c/em\u003e, vol. 11, no. September, pp. 94281\u0026ndash;94289, 2023, doi: 10.1109/ACCESS.2023.3310428.\u003c/li\u003e\n\u003cli\u003eI. Naseer, S. Akram, T. Masood, A. Jaffar, M. A. Khan, and A. Mosavi, \u0026ldquo;Performance Analysis of State-of-the-Art CNN Architectures for LUNA16,\u0026rdquo; \u003cem\u003eSensors 2022, Vol. 22, Page 4426\u003c/em\u003e, vol. 22, no. 12, p. 4426, Jun. 2022, doi: 10.3390/S22124426.\u003c/li\u003e\n\u003cli\u003eR. Ali \u003cem\u003eet al.\u003c/em\u003e, \u0026ldquo;Deep Learning for Sarcasm Identification in News Headlines,\u0026rdquo; \u003cem\u003eAppl. Sci.\u003c/em\u003e, vol. 13, no. 9, 2023, doi: 10.3390/app13095586.\u003c/li\u003e\n\u003cli\u003eI. Naseer, T. Masood, S. Akram, A. Jaffar, M. Rashid, and M. A. Iqbal, \u0026ldquo;Lung Cancer Detection Using Modified AlexNet Architecture and Support Vector Machine,\u0026rdquo; \u003cem\u003eComput. Mater. Contin.\u003c/em\u003e, vol. 74, no. 1, pp. 2039\u0026ndash;2054, 2023, doi: 10.32604/cmc.2023.032927.\u003c/li\u003e\n\u003cli\u003eW. Arshad \u003cem\u003eet al.\u003c/em\u003e, \u0026ldquo;Cancer Unveiled: A Deep Dive Into Breast Tumor Detection Using Cutting-Edge Deep Learning Models,\u0026rdquo; \u003cem\u003eIEEE Access\u003c/em\u003e, vol. 11, no. December, pp. 133804\u0026ndash;133824, 2023, doi: 10.1109/ACCESS.2023.3335604.\u003c/li\u003e\n\u003cli\u003eI. Naseer, S. Akram, T. Masood, M. Rashid, and A. Jaffar, \u0026ldquo;Lung Cancer Classification Using Modified U-Net Based Lobe Segmentation and Nodule Detection,\u0026rdquo; \u003cem\u003eIEEE Access\u003c/em\u003e, vol. 11, no. June, pp. 60279\u0026ndash;60291, 2023, doi: 10.1109/ACCESS.2023.3285821.\u003c/li\u003e\n\u003cli\u003eD. Chicco and G. Jurman, \u0026ldquo;The advantages of the Matthews correlation coefficient (MCC) over F1 score and accuracy in binary classification evaluation,\u0026rdquo; \u003cem\u003eBMC Genomics\u003c/em\u003e, vol. 21, no. 1, pp. 1\u0026ndash;13, 2020, doi: 10.1186/s12864-019-6413-7.\u003c/li\u003e\n\u003cli\u003eL. I. Kuncheva, \u0026Aacute;. Arnaiz-Gonz\u0026aacute;lez, J. F. D\u0026iacute;ez-Pastor, and I. A. D. Gunn, \u0026ldquo;Instance selection improves geometric mean accuracy: a study on imbalanced data classification,\u0026rdquo; \u003cem\u003eProg. Artif. Intell.\u003c/em\u003e, vol. 8, no. 2, pp. 215\u0026ndash;228, 2019, doi: 10.1007/s13748-019-00172-4. \u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Brain tumor, Figshare, transfer learning, CNN, optimizers, MRI","lastPublishedDoi":"10.21203/rs.3.rs-6937303/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6937303/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eOne of the most lethal diseases in the world is brain tumors. Convolutional neural networks (CNNs) and other deep learning-based methods are vital for accurate diagnosis because of their remarkable capacity for learning and prediction. These techniques are widely used in image processing, computer vision, and medical diagnostic tasks such as classification, segmentation, and object detection. In this study, we use the publicly available Figshare brain tumor dataset to compare the performance of different pre-trained deep learning models, including AlexNet, VGG16, VGG19, ResNet50, Xception, InceptionV3, DenseNet121, MobileNetV1, and EfficientNetB0, on four optimizers, such as adaptive moment estimation (Adam), stochastic gradient descent (SGD), root mean square propagation (RMSProp), and Nesterov-accelerated adaptive moment estimation (Nadam). The experimental results show that the VGG19 and EfficientNetB0 models performed exceptionally well with the Adam optimizer, achieving an overall accuracy of 99.13% and a misclassification rate of 0.87%. Additionally, receiver operating characteristic (ROC) curves were calculated, with the EfficientNetB0 model achieving an area under the curve (AUC) value of 100% for each class. It also demonstrated excellent performance on the test images, with a testing accuracy of 99.61%.\u003c/p\u003e","manuscriptTitle":"Optimizer-Aware Deep Learning for Brain Tumor Classification: A Study Using AlexNet to EfficientNetB0","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-06-25 07:29:07","doi":"10.21203/rs.3.rs-6937303/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"19ff3019-209f-4980-8c5a-be158fc956db","owner":[],"postedDate":"June 25th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-06-25T07:29:07+00:00","versionOfRecord":[],"versionCreatedAt":"2025-06-25 07:29:07","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6937303","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6937303","identity":"rs-6937303","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}