Machine Learning-based Breast Cancer Classification Using Logistic Regression and Random Forest

Machine Learning-based Breast Cancer Classification Using Logistic Regression and Random Forest | 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 Machine Learning-based Breast Cancer Classification Using Logistic Regression and Random Forest kandadi thirupathi reddy, vannala srujan This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7638704/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 Breast cancer categorization is vital for prompt identification and strategic therapy planning. In this research, a machine learning model was created utilizing the Breast Cancer Wisconsin (Diagnostic) dataset to categorize tumors as malignant or benign. The model achieved an overall accuracy of 97.36%, demonstrating strong predictive performance. For the malignant class, the model attained a precision of 0.976, recall of 0.953, and F1-score of 0.965, indicating a strong accuracy in detecting cancerous instances. For the benign class , the model achieved a precision of 0.972, recall of 0.986, and F1-score of 0.979, confirming its effectiveness in correctly classifying non-cancerous cases. The macro average F1-score was 0.972, and the weighted average F1-score was 0.974, further emphasizing equitable performance in both categories. These findings indicate that the suggested classification method offers a reliable and precise diagnostic instrument, with possible uses in clinical decision support systems for breast cancer detection. General Biochemistry Cancer Biology Artificial Intelligence and Machine Learning Cellular & Molecular Neuroscience Biophysics Breast Cancer Classification Machine Learning Diagnostic Accuracy Malignant vs. Benign Predictive Modeling Medical Diagnosis Data-driven Healthcare Classification Metrics Precision and Recall Clinical Decision Support Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Introduction Breast cancer ranks among the primary causes of cancer-related fatalities among women globally. Timely identification and precise diagnosis are essential for enhancing patient results and lowering death rates. Conventionally, diagnosis depends on techniques like mammography, biopsy, and clinical assessment. Although these methods are successful, they can be lengthy, prone to human mistakes, and reliant on the skill of healthcare practitioners. In recent years, machine learning (ML) techniques have demonstrated considerable promise in automating and improving the precision of breast cancer diagnosis. Through the examination of clinical and imaging data, ML models are capable of recognizing intricate patterns that differentiate between malignant (cancerous) and benign (non-cancerous) tumors. Among various datasets available, the Breast Cancer Wisconsin (Diagnostic) dataset is commonly utilized for investigation because of its organized characteristics obtained from fine needle aspirates (FNA) of breast tumors. This research aims to create a strong machine learning-based classification model for breast cancer utilizing this dataset. The goals are to: Classify tumors as malignant or benign with high accuracy. Assess model efficiency employing measures such as precision, recall, F1-score, and ROC-AUC. Offer a resource that can aid healthcare professionals in identifying issues at an early stage and enhance their decision-making strategies. By leveraging models such as Logistic Regression and Random Forests, this initiative seeks to illustrate those data-centric techniques can enhance conventional diagnostic practices, possibly resulting in quicker, more dependable, and understandable breast cancer assessments. Research Gap Notwithstanding considerable progress in machine learning (ML) for medical diagnostics, there are still numerous gaps present in breast cancer classification studies: Limited interpretability : Numerous highly accurate models, particularly those based on deep learning techniques; function as "black boxes," creating challenges for healthcare professionals in terms of trusting and comprehending their predictions. Small dataset reliance : Most studies rely on the Breast Cancer Wisconsin dataset, which, although commonly utilized, is quite limited and may not completely reflect the diversity of the population. Feature utilization : Certain models utilize merely a portion of the features at hand, which may lead to the underuse of crucial diagnostic data. Real-world deployment : Numerous research works concentrate on creating and evaluating models in regulated settings, paying little attention to intuitive resources for medical decision assistance or batch forecasting for several patients. These voids signify a requirement for explainable, precise, and applicable ML models that utilize all pertinent characteristics and can aid healthcare professionals in practical situations. Problem Statement Breast cancer detection continues to be a difficult endeavor because of the diversity in tumor features and dependence on human analysis of clinical information. Although machine learning algorithms have shown considerable predictive precision, there is an absence of accessible, interpretable, and comprehensive tools that can: Employ all pertinent characteristics for precise tumor categorization Offer distinct performance indicators and graphical illustrations Enable batch forecasting for numerous patient files Aid healthcare professionals in prompt and dependable identification This initiative seeks to tackle these obstacles through the creation of a robust, interactive ML-based classification system utilizing Logistic Regression and Random Forests, proficient in making both individual and batch forecasts, alongside interpretability functionalities such as feature importance and ROC visualization, thus connecting research with real-world clinical implementation Literature Review Breast cancer ranks as one of the most common cancers affecting women worldwide, and prompt identification is essential for successful treatment and enhanced survival statistics. Conventional diagnostic approaches, like mammography and biopsy, are lengthy and susceptible to personal interpretation, prompting researchers to investigate machine learning (ML) methods for automated categorization. The Breast Cancer Wisconsin (Diagnostic) dataset has been extensively utilized in ML research because of its well-organized characteristics obtained from fine needle aspirates of breast masses. Numerous studies have employed different ML algorithms, including Support Vector Machines (SVM), k-Nearest Neighbors (k-NN), Decision Trees, Random Forests, and Logistic Regression, to categorize tumors as cancerous or non-cancerous. For instance, Duaand Graff (2019) showed that Random Forests attained impressive precision (~ 97%) on the WDBC dataset, utilizing ensemble learning to minimize over fitting and enhance generalization. Likewise, Logistic Regression and SVM have proven to be successful in differentiating between malignant and benign instances, with excellent accuracy and recall, thus rendering them appropriate for clinical decision support systems. Recent advancements in deep learning, particularly convolutional neural networks (CNNs) have likewise been utilized for histopathological and imaging datasets, providing encouraging outcomes in automated tumor identification and segmentation. Nonetheless, these methods generally necessitate more extensive datasets and greater computational power than conventional ML models. Overall, the research emphasizes that machine learning models deliver dependable, effective, and repeatable results for breast cancer classification. The blend of predictive precision, clarity of interpretation, and simplicity of application renders models such as Random Forests and Logistic Regression particularly apt for aiding in early detection and clinical decision processes. Proposed System The proposed system is a framework for breast cancer classification based on machine learning crafted to help medical professionals reliably differentiate between cancerous and non-cancerous tumors. The system utilizes the Breast Cancer Wisconsin (Diagnostic) dataset, which offers 30 quantitative characteristics derived from fine needle aspirate (FNA) specimens. The main elements of the system are outlined below: 1. Data Acquisition and Preprocessing The dataset is obtained from the UCI Machine Learning Repository or directly via scikit-learn. Data preprocessing includes handling missing values, standardization of features, and dividing into training and testing groups. 2. Machine Learning Models Two models have been developed for categorization: Logistic Regression (LR) : A linear model that offers understandable coefficients to comprehend the impact of features. Random Forest (RF) : A tree-based ensemble model that identifies non-linear interactions and offers rankings for feature significance. Users can choose the model through an interactive interface and modify hyper parameters such as the quantity of trees in RF. 3. Model Training and Evaluation Models are developed using the preprocessed training data and assessed on the test set. Performance metrics include accuracy, exactness, recall, F1-measure, confusion table, and ROC-AUC. Visualizations such as feature importance plots and ROC curves enhance interpretability. Precision = TP/TP + FP 4. Interactive Prediction Interface The system provides a Streamlit-based user interface for: Manual input : Individuals can input values for chosen characteristics to receive immediate forecasts. Batch predictions : Users are able to upload CSV documents containing several patient records, and the system provides predictions along with a downloadable results file. 5. Interpretability and Insights The significance of features from Random Forest or the coefficients from Logistic Regression are illustrated to assist clinicians in comprehending which factors have the greatest impact on predictions. ROC curves offer an understanding of the balance between true positive and false positive rates, facilitating dependable decision-making assistance. Y = mode {h 1​ (x),h 2​ (x),h 3 ​(x),…,h n ​(x)} 6. Deployment The system is capable of being deployed as a web application, allowing healthcare practitioners to utilize it without needing advanced ML expertise. The interface allows interactive exploration, prediction, and visualization, connecting the divide between research frameworks and real-world clinical applications. Advantages of the Proposed System: High precision in differentiating between malignant and benign tumors Interpretable results with visual explanations Accommodates both individual and group predictions Intuitive interface designed for assisting clinical decision-making Methodology The approach for this breast cancer classification system includes multiple essential phases, ranging from data gathering to model assessment and implementation. The procedure aims to provide predictions that are precise, understandable, and applicable. 1. Data Collection The dataset used is the Breast Cancer Wisconsin (Diagnostic) dataset, obtained from the UCI Machine Learning Repository or via scikit-learn. The dataset contains 569 instances with 30 numerical features obtained from fine needle aspiration (FNA) images of breast tumors. The variable of interest is binary: Malignant (0) or Benign (1). 2. Data Preprocessing Handling Missing Values : Any absent or empty entries are verified and addressed accordingly. Feature Scaling : Standardization is utilized to standardize feature values using Standard Scaler, enhancing model convergence and effectiveness. Train-Test Split : The dataset is split into training (80%) and testing (20%) sets to assess generalization effectiveness. 3. Model Selection Two machine learning models are developed and evaluated: Logistic Regression (LR) : A straightforward model offering clear coefficients to comprehend the impact of features. Random Forest (RF) : A collection of decision trees that can identify non-linear connections and offer rankings of feature significance. Hyper parameters such as number of trees (for RF) can be modified to enhance efficiency. 4. Model Training The models utilize the preprocessed training dataset for training. Cross-validation methods can be utilized to guarantee stability and avert overfitting. The models acquire the ability to differentiate between malignant and benign tumors by utilizing 30 feature inputs. 5. Model Evaluation Performance metrics used include: Accuracy : Overall correctness of predictions Precision : Correctness of positive predictions Recall (Sensitivity) : Capacity to recognize cancerous instances F1-Score : Harmonic average of precision and recall Confusion Matrix : Depiction of actual versus anticipated labels ROC Curve & AUC : Assess the balance between the true positive rate and the false positive rate These measurements guarantee that the model is both predictively accurate and dependable for clinical application. 6. Prediction Interface A Streamlit web interface is developed to allow: Manual predictions : Users enter feature values to receive immediate tumor classification. Batch predictions : CSV upload enables the simultaneous categorization of numerous patient records. Plots of feature significance and ROC curves are offered for clarity. 7. Deployment The system is deployed as a web application, allowing healthcare providers to utilize it without needing expertise in machine learning. The interface offers visual data and forecasting results, improving user experience and clinical significance. Results and Discussion The suggested system was assessed utilizing the Breast Cancer Wisconsin (Diagnostic) dataset, which comprises 569 samples with 30 attributes. The dataset was divided into training (80%) and testing (20%) sets. Both Logistic Regression and Random Forest models were developed and evaluated, with effectiveness measured by accuracy, precision, recall, F1-score, and ROC-AUC. 1. Model Performance Metrics Table 1 Model performance metrics. Class Precision Recall F1-score Support Malignant 0.976 0.953 0.965 43 Benign 0.972 0.986 0.979 71 Accuracy 0.974 0.974 0.974 114 Macro Avg 0.974 0.970 0.972 114 Weighted Avg 0.974 0.974 0.974 114 The overall accuracy of the system is 97.37%, signifying extremely dependable classification results. The F1-score values for both cancerous (0.965) and non-cancerous (0.979) categories show a robust balance between precision and recall. The macro average F1-score (0.972) and weighted average F1-score (0.974) indicate steady performance in both categories. 2. Confusion Matrix Analysis The confusion matrix shows that: The majority of cancerous cases were accurately recognized, with only a small number of false negatives. Non-threatening cases were also accurately categorized, reducing false positives. This demonstrates the system’s ability to reduce the misidentification of essential cancerous tumors, which is essential for medical use. 3. ROC-AUC The ROC curve analysis yielded a high AUC, demonstrating that the models are successful in differentiating between malignant and benign tumors. Both models exhibit outstanding discrimination, with Random Forest marginally exceeding Logistic Regression in sensitivity to malignant instances. 4. Feature Importance Random Forest feature significance evaluation indicates that attributes such as average radius, average concavity, and maximum perimeter are very significant in forecasting tumor aggressiveness. Logistic Regression coefficients also emphasize the significance of essential features, improving clarity for clinical decision-making. 5. Discussion The system demonstrates high predictive accuracy and robust performance on unseen test data. Batch prediction capability facilitates the swift assessment of numerous patient records, which is beneficial for practical clinical processes. Visual interpretability tools, including feature importance plots and ROC curves, offer clarity on model choices, enhancing confidence for healthcare professionals. In relation to earlier research utilizing the same dataset, the system demonstrates similar or slightly enhanced performance, confirming its efficiency. Limitations: The size of the dataset is somewhat limited, and practical application in the real world may necessitate larger and more varied datasets to ensure generalization across different groups. The existing model does not include imaging information, which could enhance prediction precision when integrated with clinical characteristics. Conclusion from Results The suggested breast cancer categorization system is accurate, interpretable, and clinically relevant, rendering it appropriate as a decision-making aid for prompt identification and therapy preparation. Conclusion This study presented an algorithm-driven breast cancer categorization system using the Breast Cancer Wisconsin (Diagnostic) dataset. Two models, Logistic Regression and Random Forest, were put into practice and assessed for their capability to categorize tumors as cancerous or non-cancerous. The system achieved an overall accuracy of 97.37%, achieving high precision, recall, and F1 scores for both categories. The ROC - AUC analysis and feature importance visualization showcased the strength and clarity of the models, enabling healthcare professionals to comprehend which factors have the greatest impact on predictions. The proposed system provides a user-friendly interface for both individual patient and group predictions, closing the divide between research-based models and real-world clinical uses. Its excellent performance, clarity of understanding, and user-friendly nature render it a valuable resource for early detection and decision support in breast cancer diagnosis. Future work could focus on: Incorporating larger and more diverse datasets to improve generalization. Integrating imaging data with clinical characteristics to improve forecasting precision. Expanding the system with advanced deep learning models while maintaining interpretability. Overall, this study demonstrates that machine learning algorithms can greatly support precise, effective, and understandable breast cancer detection, aiding prompt clinical actions and enhancing patient results. References Dua, D., & Graff, C. (2019). UCI Machine Learning Repository . Irvine, CA: University of California, School of Information and Computer Science. Retrieved from http://archive.ics.uci.edu/ml Street, W. N., Wolberg, W. H., & Mangasarian, O. L. (1993). Nuclear feature extraction for breast tumor diagnosis. IS&T/SPIE 1993 International Symposium on Electronic Imaging: Science and Technology , 1905, 861–870. Wolberg, W. H., & Mangasarian, O. L. (1990). Multisurface method of pattern separation for medical diagnosis applied to breast cytology. Proceedings of the National Academy of Sciences , 87(23), 9193–9196. https://doi.org/10.1073/pnas.87.23.9193 Sahan, S., Polat, K., Kodaz, H., & Günes, S. (2007). A new hybrid method based on fuzzy–artificial immune system and k-nn algorithm for breast cancer diagnosis. Computers in Biology and Medicine , 37(3), 415–423. https://doi.org/10.1016/j.compbiomed.2006.05.003 Shankarlingam G, Reddy KT. (2023) Predicting a Small Cap Company Stock Price using Python with Best Accuracy Rate: How the Data Science Working for Predictions and Accuracy Rate. Indian Journal of Science and Technology. 16(48): 4620- 4623. https://doi.org/10.17485/IJST/v16i48.2793 thirupathi. A Streamlit-powered System for Simulating and Visualizing Qubit States, 11 September 2025, PREPRINT (Version 1) available at Research Square [https://doi.org/10.21203/rs.3.rs-7579530/v1] Kandadi, Thirupathi and Shankarlingam, G., Analysis Report of Traders Losing Money in Trading Platforms: A Sebi Report (October 26, 2024). ssrn, Available at SSRN: https://ssrn.com/abstract=5000105 or http://dx.doi.org/10.2139/ssrn.5000105 Kandadi, Thirupathi and Shankarlingam, G., DRAWBACKS OF LSTM ALGORITHM: A CASE STUDY (January 01, 2025). Available at SSRN: https://ssrn.com/abstract=5080605 or http://dx.doi.org/10.2139/ssrn.5080605 Kandadi, Thirupathi, IMPLEMENTATION OF DICHOTOMISER 3 ALGORITHM WITH DECISION TREE FOR MAKING DECISIONS IN MACHINE LEARNING (April 07, 2025). Available at SSRN: https://ssrn.com/abstract=5207811 or http://dx.doi.org/10.2139/ssrn.5207811 Kandadi Thirupathi Reddy. A FRAMEWORK TO PREDICTING STARTUP SUCCESS GROWTH WITH MULTIPLE ALGORITHMS. TechRxiv. September 10, 2025. DOI: 10.36227/techrxiv.175751434.42657764/v1 Additional Declarations The authors declare no competing interests. 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-7638704","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":516432654,"identity":"5faed027-d12f-4617-90ee-004f6018b5d9","order_by":0,"name":"kandadi thirupathi reddy","email":"data:image/png;base64,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","orcid":"https://orcid.org/0009-0007-0835-0581","institution":"chaitanya deemed to be university","correspondingAuthor":true,"prefix":"","firstName":"kandadi","middleName":"thirupathi","lastName":"reddy","suffix":""},{"id":516432655,"identity":"29fe5747-3f0b-40ca-8003-56fc20338937","order_by":1,"name":"vannala srujan","email":"","orcid":"https://orcid.org/0009-0009-3481-9421","institution":"chaitanya deemed to be university","correspondingAuthor":false,"prefix":"","firstName":"vannala","middleName":"","lastName":"srujan","suffix":""}],"badges":[],"createdAt":"2025-09-17 09:55:58","currentVersionCode":1,"declarations":{"humanSubjects":true,"vertebrateSubjects":true,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":true,"humanSubjectConsent":true,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":true},"doi":"10.21203/rs.3.rs-7638704/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7638704/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":91962122,"identity":"44ea4554-5595-4e8f-91e1-6b6240235d9a","added_by":"auto","created_at":"2025-09-23 07:54:24","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":250958,"visible":true,"origin":"","legend":"","description":"","filename":"article.docx","url":"https://assets-eu.researchsquare.com/files/rs-7638704/v1/bc541cdfad2925e408ec781a.docx"},{"id":91962136,"identity":"1551d19a-f485-4dc6-a419-ea470eb0e7d4","added_by":"auto","created_at":"2025-09-23 07:54:26","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":342,"visible":true,"origin":"","legend":"","description":"","filename":"rs7638704.json","url":"https://assets-eu.researchsquare.com/files/rs-7638704/v1/2759b9c21627a948d54b5166.json"},{"id":91962164,"identity":"69fa5f92-0cbf-41c0-90ab-b6ae2b54c884","added_by":"auto","created_at":"2025-09-23 07:54:27","extension":"xml","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":51308,"visible":true,"origin":"","legend":"","description":"","filename":"rs76387040enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7638704/v1/869c9797d9812db4382848a3.xml"},{"id":91962170,"identity":"d4a19630-a41d-41da-bd8c-1b8039db3e99","added_by":"auto","created_at":"2025-09-23 07:54:28","extension":"jpeg","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":27283,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7638704/v1/e18ad1e68dc888260cedbc63.jpeg"},{"id":91962129,"identity":"c7544daa-5496-410e-bb8e-50df04cc1170","added_by":"auto","created_at":"2025-09-23 07:54:25","extension":"jpeg","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":50432,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7638704/v1/6d28f4bd800ec4ff4f88f1cc.jpeg"},{"id":91962118,"identity":"5df74cee-5711-4633-9c2f-be0dc4d9c50b","added_by":"auto","created_at":"2025-09-23 07:54:23","extension":"jpeg","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":8617,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage3.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7638704/v1/d34c6bcfaea9c12895f0a30e.jpeg"},{"id":91962169,"identity":"cf3872f4-f864-4c92-8dd9-bc55eff0f60f","added_by":"auto","created_at":"2025-09-23 07:54:28","extension":"jpeg","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":33617,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage4.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7638704/v1/b56a0ad31439e5a7f2b0b228.jpeg"},{"id":91962139,"identity":"2c88f53a-ccd4-44ed-9fed-6266adff5f13","added_by":"auto","created_at":"2025-09-23 07:54:27","extension":"jpeg","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":42320,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage5.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7638704/v1/cb33b03962ee5e9e89cd334f.jpeg"},{"id":91962176,"identity":"d069922e-c672-449d-a82e-8015bff632b8","added_by":"auto","created_at":"2025-09-23 07:54:29","extension":"jpeg","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":51334,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage6.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7638704/v1/ee9e6bc012e32a13394398e2.jpeg"},{"id":91962107,"identity":"023118e1-6b62-4ea4-a365-8efc636a3092","added_by":"auto","created_at":"2025-09-23 07:54:22","extension":"png","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":6173,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7638704/v1/d60294f5997f4f6a7334ceaa.png"},{"id":91962137,"identity":"3fa2fd1e-4531-4200-826f-86b549013386","added_by":"auto","created_at":"2025-09-23 07:54:26","extension":"png","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":14274,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7638704/v1/202ec80f812608b50900e554.png"},{"id":91962111,"identity":"71d1f772-5292-46e7-8c2c-f2db2fc7b419","added_by":"auto","created_at":"2025-09-23 07:54:22","extension":"png","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":7280,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7638704/v1/cc80e24592c602d404f26de6.png"},{"id":91962119,"identity":"eb59a043-94ca-4e0f-b91c-759e356e9396","added_by":"auto","created_at":"2025-09-23 07:54:23","extension":"png","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":13486,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-7638704/v1/bd846b925ee45e1f35a1b384.png"},{"id":91962187,"identity":"fba8200c-88a1-4ad8-a1eb-ec8720f73b1c","added_by":"auto","created_at":"2025-09-23 07:54:29","extension":"png","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":11488,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-7638704/v1/f9bebd3c7ce9dfb26499fc1e.png"},{"id":91962143,"identity":"a756d0d2-4de8-4159-ad4d-46b9122136f2","added_by":"auto","created_at":"2025-09-23 07:54:27","extension":"png","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":11889,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-7638704/v1/82a521d3082faa7f34fe5cb0.png"},{"id":91962113,"identity":"8a28b56d-55cd-4c33-b1b3-0d4895772f6b","added_by":"auto","created_at":"2025-09-23 07:54:23","extension":"xml","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":48575,"visible":true,"origin":"","legend":"","description":"","filename":"rs76387040structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7638704/v1/f0bc26375a2595ff9b0b2fb5.xml"},{"id":91962175,"identity":"4fbf8343-da26-4504-8ddd-1bce994576ef","added_by":"auto","created_at":"2025-09-23 07:54:28","extension":"html","order_by":16,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":60017,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7638704/v1/d3ab4f5d8e8eed4a899a00af.html"},{"id":91962165,"identity":"0b138263-59f2-46de-a11b-905080c20e83","added_by":"auto","created_at":"2025-09-23 07:54:27","extension":"jpeg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":27283,"visible":true,"origin":"","legend":"\u003cp\u003eManual prediction through dataset values.\u003c/p\u003e","description":"","filename":"image1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7638704/v1/8e7663886dee6d802fd75cff.jpeg"},{"id":91962128,"identity":"9f981b9e-9737-42cd-a311-219aab394632","added_by":"auto","created_at":"2025-09-23 07:54:25","extension":"jpeg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":50432,"visible":true,"origin":"","legend":"\u003cp\u003eDataset upload option and results display and download option.\u003c/p\u003e","description":"","filename":"image2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7638704/v1/d89378a92098387c58eade68.jpeg"},{"id":91963273,"identity":"d8acbe52-7cb0-4818-8eed-2ceed4f10614","added_by":"auto","created_at":"2025-09-23 08:02:27","extension":"jpeg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":8617,"visible":true,"origin":"","legend":"\u003cp\u003eChoosing algorithm and test size selection.\u003c/p\u003e","description":"","filename":"image3.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7638704/v1/1fed508bb920ef828a325fef.jpeg"},{"id":91962123,"identity":"54e6ccbc-2824-4aad-9c75-f5b229b76f10","added_by":"auto","created_at":"2025-09-23 07:54:24","extension":"jpeg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":33617,"visible":true,"origin":"","legend":"\u003cp\u003eModel performance statistics.\u003c/p\u003e","description":"","filename":"image4.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7638704/v1/921c85e31d20b3c9776dd972.jpeg"},{"id":91962168,"identity":"b6f00713-a260-4b20-bd2a-d7676e0cf425","added_by":"auto","created_at":"2025-09-23 07:54:27","extension":"jpeg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":42320,"visible":true,"origin":"","legend":"\u003cp\u003eROC-AUC curve for FP and TFP rate.\u003c/p\u003e","description":"","filename":"image5.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7638704/v1/f16f3186ccbf113462d05e5f.jpeg"},{"id":91962174,"identity":"0571438a-1371-4c3d-8441-e1f4f773ce6b","added_by":"auto","created_at":"2025-09-23 07:54:28","extension":"jpeg","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":51334,"visible":true,"origin":"","legend":"\u003cp\u003eFeature importance coefficients.\u003c/p\u003e","description":"","filename":"image6.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7638704/v1/77dc66ceb10204b4c0e05127.jpeg"},{"id":91963282,"identity":"a9d4bfd0-0a76-451e-beaa-a7b1aedf7a1b","added_by":"auto","created_at":"2025-09-23 08:02:32","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1039190,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7638704/v1/09c951ab-bbbd-44a7-83d7-cfcbb3b3e22a.pdf"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003e\u003cstrong\u003eMachine Learning-based Breast Cancer Classification Using Logistic Regression and Random Forest\u003c/strong\u003e\u003c/p\u003e","fulltext":[{"header":"Introduction","content":"\u003cp\u003eBreast cancer ranks among the primary causes of cancer-related fatalities among women globally. Timely identification and precise diagnosis are essential for enhancing patient results and lowering death rates. Conventionally, diagnosis depends on techniques like mammography, biopsy, and clinical assessment. Although these methods are successful, they can be lengthy, prone to human mistakes, and reliant on the skill of healthcare practitioners.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eIn recent years, machine learning (ML) techniques have demonstrated considerable promise in automating and improving the precision of breast cancer diagnosis. Through the examination of clinical and imaging data, ML models are capable of recognizing intricate patterns that differentiate between malignant (cancerous) and benign (non-cancerous) tumors. Among various datasets available, the Breast Cancer Wisconsin (Diagnostic) dataset is commonly utilized for investigation because of its organized characteristics obtained from fine needle aspirates (FNA) of breast tumors.\u003c/p\u003e\u003cp\u003eThis research aims to create a strong machine learning-based classification model for breast cancer utilizing this dataset. The goals are to:\u003c/p\u003e\u003cp\u003e\u003col\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003eClassify tumors as malignant or benign with high accuracy.\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003eAssess model efficiency employing measures such as precision, recall, F1-score, and ROC-AUC.\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003eOffer a resource that can aid healthcare professionals in identifying issues at an early stage and enhance their decision-making strategies.\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003c/ol\u003e\u003c/p\u003e\u003cp\u003eBy leveraging models such as Logistic Regression and Random Forests, this initiative seeks to illustrate those data-centric techniques can enhance conventional diagnostic practices, possibly resulting in quicker, more dependable, and understandable breast cancer assessments.\u003c/p\u003e\n\u003ch3\u003eResearch Gap\u003c/h3\u003e\n\u003cp\u003eNotwithstanding considerable progress in machine learning (ML) for medical diagnostics, there are still numerous gaps present in breast cancer classification studies:\u003c/p\u003e\u003cp\u003e\u003col\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003e\u003cb\u003eLimited interpretability\u003c/b\u003e: Numerous highly accurate models, particularly those based on deep learning techniques; function as \"black boxes,\" creating challenges for healthcare professionals in terms of trusting and comprehending their predictions.\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003e\u003cb\u003eSmall dataset reliance\u003c/b\u003e: Most studies rely on the Breast Cancer Wisconsin dataset, which, although commonly utilized, is quite limited and may not completely reflect the diversity of the population.\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003e\u003cb\u003eFeature utilization\u003c/b\u003e: Certain models utilize merely a portion of the features at hand, which may lead to the underuse of crucial diagnostic data.\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003e\u003cb\u003eReal-world deployment\u003c/b\u003e: Numerous research works concentrate on creating and evaluating models in regulated settings, paying little attention to intuitive resources for medical decision assistance or batch forecasting for several patients.\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003c/ol\u003e\u003c/p\u003e\u003cp\u003eThese voids signify a requirement for explainable, precise, and applicable ML models that utilize all pertinent characteristics and can aid healthcare professionals in practical situations.\u003c/p\u003e\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003eProblem Statement\u003c/h2\u003e\u003cp\u003eBreast cancer detection continues to be a difficult endeavor because of the diversity in tumor features and dependence on human analysis of clinical information. Although machine learning algorithms have shown considerable predictive precision, there is an absence of accessible, interpretable, and comprehensive tools that can:\u003c/p\u003e\u003cp\u003e\u003cul\u003e\u003cli\u003e\u003cp\u003eEmploy all pertinent characteristics for precise tumor categorization\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eOffer distinct performance indicators and graphical illustrations\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eEnable batch forecasting for numerous patient files\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eAid healthcare professionals in prompt and dependable identification\u003c/p\u003e\u003c/li\u003e\u003c/ul\u003e\u003c/p\u003e\u003cp\u003eThis initiative seeks to tackle these obstacles through the creation of a robust, interactive ML-based classification system utilizing Logistic Regression and Random Forests, proficient in making both individual and batch forecasts, alongside interpretability functionalities such as feature importance and ROC visualization, thus connecting research with real-world clinical implementation\u003c/p\u003e\u003c/div\u003e"},{"header":"Literature Review","content":"\u003cp\u003eBreast cancer ranks as one of the most common cancers affecting women worldwide, and prompt identification is essential for successful treatment and enhanced survival statistics. Conventional diagnostic approaches, like mammography and biopsy, are lengthy and susceptible to personal interpretation, prompting researchers to investigate machine learning (ML) methods for automated categorization.\u003c/p\u003e\u003cp\u003eThe Breast Cancer Wisconsin (Diagnostic) dataset has been extensively utilized in ML research because of its well-organized characteristics obtained from fine needle aspirates of breast masses. Numerous studies have employed different ML algorithms, including Support Vector Machines (SVM), k-Nearest Neighbors (k-NN), Decision Trees, Random Forests, and Logistic Regression, to categorize tumors as cancerous or non-cancerous.\u003c/p\u003e\u003cp\u003eFor instance, Duaand Graff (2019) showed that Random Forests attained impressive precision (~\u0026thinsp;97%) on the WDBC dataset, utilizing ensemble learning to minimize over fitting and enhance generalization. Likewise, Logistic Regression and SVM have proven to be successful in differentiating between malignant and benign instances, with excellent accuracy and recall, thus rendering them appropriate for clinical decision support systems.\u003c/p\u003e\u003cp\u003eRecent advancements in deep learning, particularly convolutional neural networks (CNNs) have likewise been utilized for histopathological and imaging datasets, providing encouraging outcomes in automated tumor identification and segmentation. Nonetheless, these methods generally necessitate more extensive datasets and greater computational power than conventional ML models.\u003c/p\u003e\u003cp\u003eOverall, the research emphasizes that machine learning models deliver dependable, effective, and repeatable results for breast cancer classification. The blend of predictive precision, clarity of interpretation, and simplicity of application renders models such as Random Forests and Logistic Regression particularly apt for aiding in early detection and clinical decision processes.\u003c/p\u003e\n\u003ch3\u003eProposed System\u003c/h3\u003e\n\u003cp\u003eThe proposed system is a framework for breast cancer classification based on machine learning crafted to help medical professionals reliably differentiate between cancerous and non-cancerous tumors. The system utilizes the Breast Cancer Wisconsin (Diagnostic) dataset, which offers 30 quantitative characteristics derived from fine needle aspirate (FNA) specimens. The main elements of the system are outlined below:\u003c/p\u003e\n\u003ch3\u003e1. Data Acquisition and Preprocessing\u003c/h3\u003e\n\u003cp\u003e\u003cul\u003e\u003cli\u003e\u003cp\u003eThe dataset is obtained from the UCI Machine Learning Repository or directly via scikit-learn.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eData preprocessing includes handling missing values, standardization of features, and dividing into training and testing groups.\u003c/p\u003e\u003c/li\u003e\u003c/ul\u003e\u003c/p\u003e\n\u003ch3\u003e2. Machine Learning Models\u003c/h3\u003e\n\u003cp\u003e\u003cul\u003e\u003cli\u003e\u003cp\u003eTwo models have been developed for categorization:\u003c/p\u003e\u003cp\u003e\u003col\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003e\u003cb\u003eLogistic Regression (LR)\u003c/b\u003e: A linear model that offers understandable coefficients to comprehend the impact of features.\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003e\u003cb\u003eRandom Forest (RF)\u003c/b\u003e: A tree-based ensemble model that identifies non-linear interactions and offers rankings for feature significance.\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003c/ol\u003e\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eUsers can choose the model through an interactive interface and modify hyper parameters such as the quantity of trees in RF.\u003c/p\u003e\u003c/li\u003e\u003c/ul\u003e\u003c/p\u003e\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e\u003ch2\u003e3. Model Training and Evaluation\u003c/h2\u003e\u003cp\u003e\u003cul\u003e\u003cli\u003e\u003cp\u003eModels are developed using the preprocessed training data and assessed on the test set.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003ePerformance metrics include accuracy, exactness, recall, F1-measure, confusion table, and ROC-AUC.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eVisualizations such as feature importance plots and ROC curves enhance interpretability.\u003c/p\u003e\u003c/li\u003e\u003c/ul\u003e\u003c/p\u003e\u003c/div\u003e\n\u003ch3\u003ePrecision = TP/TP + FP\u003c/h3\u003e\n\u003cdiv id=\"Sec10\" class=\"Section2\"\u003e\u003ch2\u003e4. Interactive Prediction Interface\u003c/h2\u003e\u003cp\u003e\u003cul\u003e\u003cli\u003e\u003cp\u003eThe system provides a Streamlit-based user interface for:\u003c/p\u003e\u003cp\u003e\u003cul\u003e\u003cli\u003e\u003cp\u003e\u003cb\u003eManual input\u003c/b\u003e: Individuals can input values for chosen characteristics to receive immediate forecasts.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003e\u003cb\u003eBatch predictions\u003c/b\u003e: Users are able to upload CSV documents containing several patient records, and the system provides predictions along with a downloadable results file.\u003c/p\u003e\u003c/li\u003e\u003c/ul\u003e\u003c/p\u003e\u003c/li\u003e\u003c/ul\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e\u003ch2\u003e5. Interpretability and Insights\u003c/h2\u003e\u003cp\u003e\u003cul\u003e\u003cli\u003e\u003cp\u003eThe significance of features from Random Forest or the coefficients from Logistic Regression are illustrated to assist clinicians in comprehending which factors have the greatest impact on predictions.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eROC curves offer an understanding of the balance between true positive and false positive rates, facilitating dependable decision-making assistance.\u003c/p\u003e\u003c/li\u003e\u003c/ul\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec12\" class=\"Section2\"\u003e\u003ch2\u003eY\u0026thinsp;=\u0026thinsp;mode {h\u003csub\u003e1​\u003c/sub\u003e(x),h\u003csub\u003e2​\u003c/sub\u003e(x),h\u003csub\u003e3\u003c/sub\u003e​(x),\u0026hellip;,h\u003csub\u003en\u003c/sub\u003e​(x)}\u003c/h2\u003e\u003cdiv id=\"Sec13\" class=\"Section3\"\u003e\u003ch2\u003e6. Deployment\u003c/h2\u003e\u003cp\u003e\u003cul\u003e\u003cli\u003e\u003cp\u003eThe system is capable of being deployed as a web application, allowing healthcare practitioners to utilize it without needing advanced ML expertise.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eThe interface allows interactive exploration, prediction, and visualization, connecting the divide between research frameworks and real-world clinical applications.\u003c/p\u003e\u003c/li\u003e\u003c/ul\u003e\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv id=\"Sec14\" class=\"Section2\"\u003e\u003ch2\u003eAdvantages of the Proposed System:\u003c/h2\u003e\u003cp\u003e\u003cul\u003e\u003cli\u003e\u003cp\u003eHigh precision in differentiating between malignant and benign tumors\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eInterpretable results with visual explanations\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eAccommodates both individual and group predictions\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eIntuitive interface designed for assisting clinical decision-making\u003c/p\u003e\u003c/li\u003e\u003c/ul\u003e\u003c/p\u003e\u003c/div\u003e"},{"header":"Methodology","content":"\u003cdiv id=\"Sec15\" class=\"Section2\"\u003e\u003cp\u003eThe approach for this breast cancer classification system includes multiple essential phases, ranging from data gathering to model assessment and implementation. The procedure aims to provide predictions that are precise, understandable, and applicable.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec16\" class=\"Section2\"\u003e\u003ch2\u003e1. Data Collection\u003c/h2\u003e\u003cp\u003e\u003cul\u003e\u003cli\u003e\u003cp\u003eThe dataset used is the Breast Cancer Wisconsin (Diagnostic) dataset, obtained from the UCI Machine Learning Repository or via scikit-learn.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eThe dataset contains 569 instances with 30 numerical features obtained from fine needle aspiration (FNA) images of breast tumors.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eThe variable of interest is binary: Malignant (0) or Benign (1).\u003c/p\u003e\u003c/li\u003e\u003c/ul\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec17\" class=\"Section2\"\u003e\u003ch2\u003e2. Data Preprocessing\u003c/h2\u003e\u003cp\u003e\u003cul\u003e\u003cli\u003e\u003cp\u003e\u003cb\u003eHandling Missing Values\u003c/b\u003e: Any absent or empty entries are verified and addressed accordingly.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003e\u003cb\u003eFeature Scaling\u003c/b\u003e: Standardization is utilized to standardize feature values using Standard Scaler, enhancing model convergence and effectiveness.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003e\u003cb\u003eTrain-Test Split\u003c/b\u003e: The dataset is split into training (80%) and testing (20%) sets to assess generalization effectiveness.\u003c/p\u003e\u003c/li\u003e\u003c/ul\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec18\" class=\"Section2\"\u003e\u003ch2\u003e3. Model Selection\u003c/h2\u003e\u003cp\u003e\u003cul\u003e\u003cli\u003e\u003cp\u003eTwo machine learning models are developed and evaluated:\u003c/p\u003e\u003cp\u003e\u003col\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003e\u003cb\u003eLogistic Regression (LR)\u003c/b\u003e: A straightforward model offering clear coefficients to comprehend the impact of features.\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003e\u003cb\u003eRandom Forest (RF)\u003c/b\u003e: A collection of decision trees that can identify non-linear connections and offer rankings of feature significance.\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003c/ol\u003e\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eHyper parameters such as number of trees (for RF) can be modified to enhance efficiency.\u003c/p\u003e\u003c/li\u003e\u003c/ul\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec19\" class=\"Section2\"\u003e\u003ch2\u003e4. Model Training\u003c/h2\u003e\u003cp\u003e\u003cul\u003e\u003cli\u003e\u003cp\u003eThe models utilize the preprocessed training dataset for training.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eCross-validation methods can be utilized to guarantee stability and avert overfitting.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eThe models acquire the ability to differentiate between malignant and benign tumors by utilizing 30 feature inputs.\u003c/p\u003e\u003c/li\u003e\u003c/ul\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec20\" class=\"Section2\"\u003e\u003ch2\u003e5. Model Evaluation\u003c/h2\u003e\u003cp\u003e\u003cul\u003e\u003cli\u003e\u003cp\u003ePerformance metrics used include:\u003c/p\u003e\u003cp\u003e\u003cul\u003e\u003cli\u003e\u003cp\u003e\u003cb\u003eAccuracy\u003c/b\u003e: Overall correctness of predictions\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003e\u003cb\u003ePrecision\u003c/b\u003e: Correctness of positive predictions\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003e\u003cb\u003eRecall (Sensitivity)\u003c/b\u003e: Capacity to recognize cancerous instances\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003e\u003cb\u003eF1-Score\u003c/b\u003e: Harmonic average of precision and recall\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003e\u003cb\u003eConfusion Matrix\u003c/b\u003e: Depiction of actual versus anticipated labels\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003e\u003cb\u003eROC Curve \u0026amp; AUC\u003c/b\u003e: Assess the balance between the true positive rate and the false positive rate\u003c/p\u003e\u003c/li\u003e\u003c/ul\u003e\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eThese measurements guarantee that the model is both predictively accurate and dependable for clinical application.\u003c/p\u003e\u003c/li\u003e\u003c/ul\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec21\" class=\"Section2\"\u003e\u003ch2\u003e6. Prediction Interface\u003c/h2\u003e\u003cp\u003e\u003cul\u003e\u003cli\u003e\u003cp\u003eA \u003cb\u003eStreamlit web interface\u003c/b\u003e is developed to allow:\u003c/p\u003e\u003cp\u003e\u003cul\u003e\u003cli\u003e\u003cp\u003e\u003cb\u003eManual predictions\u003c/b\u003e: Users enter feature values to receive immediate tumor classification.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003e\u003cb\u003eBatch predictions\u003c/b\u003e: CSV upload enables the simultaneous categorization of numerous patient records.\u003c/p\u003e\u003c/li\u003e\u003c/ul\u003e\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003ePlots of feature significance and ROC curves are offered for clarity.\u003c/p\u003e\u003c/li\u003e\u003c/ul\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec22\" class=\"Section2\"\u003e\u003ch2\u003e7. Deployment\u003c/h2\u003e\u003cp\u003e\u003cul\u003e\u003cli\u003e\u003cp\u003eThe system is deployed as a web application, allowing healthcare providers to utilize it without needing expertise in machine learning.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eThe interface offers visual data and forecasting results, improving user experience and clinical significance.\u003c/p\u003e\u003c/li\u003e\u003c/ul\u003e\u003c/p\u003e\u003c/div\u003e"},{"header":"Results and Discussion","content":"\u003cp\u003eThe suggested system was assessed utilizing the Breast Cancer Wisconsin (Diagnostic) dataset, which comprises 569 samples with 30 attributes. The dataset was divided into training (80%) and testing (20%) sets. Both Logistic Regression and Random Forest models were developed and evaluated, with effectiveness measured by accuracy, precision, recall, F1-score, and ROC-AUC.\u003c/p\u003e\u003cp\u003e\u003cb\u003e1. Model Performance Metrics\u003c/b\u003e\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\u003eModel performance metrics.\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=\"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\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eClass\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003ePrecision\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eRecall\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eF1-score\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eSupport\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMalignant\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.976\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.953\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.965\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e43\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eBenign\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.972\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.986\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.979\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e71\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eAccuracy\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e\u003cb\u003e0.974\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e0.974\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e0.974\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e114\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMacro Avg\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.974\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.970\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.972\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e114\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eWeighted Avg\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.974\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.974\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.974\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e114\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\u003cul\u003e\u003cli\u003e\u003cp\u003eThe overall accuracy of the system is 97.37%, signifying extremely dependable classification results.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eThe F1-score values for both cancerous (0.965) and non-cancerous (0.979) categories show a robust balance between precision and recall.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eThe macro average F1-score (0.972) and weighted average F1-score (0.974) indicate steady performance in both categories.\u003c/p\u003e\u003c/li\u003e\u003c/ul\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003e2. Confusion Matrix Analysis\u003c/b\u003e\u003c/p\u003e\u003cp\u003eThe confusion matrix shows that:\u003c/p\u003e\u003cp\u003e\u003cul\u003e\u003cli\u003e\u003cp\u003eThe majority of cancerous cases were accurately recognized, with only a small number of false negatives.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eNon-threatening cases were also accurately categorized, reducing false positives.\u003c/p\u003e\u003c/li\u003e\u003c/ul\u003e\u003c/p\u003e\u003cp\u003eThis demonstrates the system\u0026rsquo;s ability to reduce the misidentification of essential cancerous tumors, which is essential for medical use.\u003c/p\u003e\u003cdiv id=\"Sec24\" class=\"Section2\"\u003e\u003ch2\u003e3. ROC-AUC\u003c/h2\u003e\u003cp\u003e\u003cul\u003e\u003cli\u003e\u003cp\u003eThe ROC curve analysis yielded a high AUC, demonstrating that the models are successful in differentiating between malignant and benign tumors.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eBoth models exhibit outstanding discrimination, with Random Forest marginally exceeding Logistic Regression in sensitivity to malignant instances.\u003c/p\u003e\u003c/li\u003e\u003c/ul\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cdiv id=\"Sec25\" class=\"Section3\"\u003e\u003ch2\u003e4. Feature Importance\u003c/h2\u003e\u003cp\u003e\u003cul\u003e\u003cli\u003e\u003cp\u003eRandom Forest feature significance evaluation indicates that attributes such as average radius, average concavity, and maximum perimeter are very significant in forecasting tumor aggressiveness.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eLogistic Regression coefficients also emphasize the significance of essential features, improving clarity for clinical decision-making.\u003c/p\u003e\u003c/li\u003e\u003c/ul\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003e5. Discussion\u003c/b\u003e\u003c/p\u003e\u003cp\u003e\u003cul\u003e\u003cli\u003e\u003cp\u003eThe system demonstrates high predictive accuracy and robust performance on unseen test data.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eBatch prediction capability facilitates the swift assessment of numerous patient records, which is beneficial for practical clinical processes.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eVisual interpretability tools, including feature importance plots and ROC curves, offer clarity on model choices, enhancing confidence for healthcare professionals.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eIn relation to earlier research utilizing the same dataset, the system demonstrates similar or slightly enhanced performance, confirming its efficiency.\u003c/p\u003e\u003c/li\u003e\u003c/ul\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec26\" class=\"Section3\"\u003e\u003ch2\u003eLimitations:\u003c/h2\u003e\u003cp\u003e\u003cul\u003e\u003cli\u003e\u003cp\u003eThe size of the dataset is somewhat limited, and practical application in the real world may necessitate larger and more varied datasets to ensure generalization across different groups.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eThe existing model does not include imaging information, which could enhance prediction precision when integrated with clinical characteristics.\u003c/p\u003e\u003c/li\u003e\u003c/ul\u003e\u003c/p\u003e\u003cp\u003e\u003cstrong\u003eConclusion from Results\u003c/strong\u003e\u003cp\u003eThe suggested breast cancer categorization system is accurate, interpretable, and clinically relevant, rendering it appropriate as a decision-making aid for prompt identification and therapy preparation.\u003c/p\u003e\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e"},{"header":"Conclusion","content":"\u003cp\u003eThis study presented an algorithm-driven breast cancer categorization system using the Breast Cancer Wisconsin (Diagnostic) dataset. Two models, Logistic Regression and Random Forest, were put into practice and assessed for their capability to categorize tumors as cancerous or non-cancerous.\u003c/p\u003e\u003cp\u003eThe system achieved an overall accuracy of 97.37%, achieving high precision, recall, and F1 scores for both categories. The ROC\u003cb\u003e-\u003c/b\u003eAUC analysis and feature importance visualization showcased the strength and clarity of the models, enabling healthcare professionals to comprehend which factors have the greatest impact on predictions.\u003c/p\u003e\u003cp\u003eThe proposed system provides a user-friendly interface for both individual patient and group predictions, closing the divide between research-based models and real-world clinical uses. Its excellent performance, clarity of understanding, and user-friendly nature render it a valuable resource for early detection and decision support in breast cancer diagnosis.\u003c/p\u003e\u003cp\u003e\u003cb\u003eFuture work\u003c/b\u003e could focus on:\u003c/p\u003e\u003cp\u003e\u003cul\u003e\u003cli\u003e\u003cp\u003eIncorporating larger and more diverse datasets to improve generalization.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eIntegrating imaging data with clinical characteristics to improve forecasting precision.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eExpanding the system with advanced deep learning models while maintaining interpretability.\u003c/p\u003e\u003c/li\u003e\u003c/ul\u003e\u003c/p\u003e\u003cp\u003eOverall, this study demonstrates that machine learning algorithms can greatly support precise, effective, and understandable breast cancer detection, aiding prompt clinical actions and enhancing patient results.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eDua, D., \u0026amp; Graff, C. (2019). \u003cem\u003eUCI Machine Learning Repository\u003c/em\u003e. Irvine, CA: University of California, School of Information and Computer Science. Retrieved from http://archive.ics.uci.edu/ml\u003c/li\u003e\n\u003cli\u003eStreet, W. N., Wolberg, W. H., \u0026amp; Mangasarian, O. L. (1993). Nuclear feature extraction for breast tumor diagnosis. \u003cem\u003eIS\u0026amp;T/SPIE 1993 International Symposium on Electronic Imaging: Science and Technology\u003c/em\u003e, 1905, 861\u0026ndash;870.\u003c/li\u003e\n\u003cli\u003eWolberg, W. H., \u0026amp; Mangasarian, O. L. (1990). Multisurface method of pattern separation for medical diagnosis applied to breast cytology. \u003cem\u003eProceedings of the National Academy of Sciences\u003c/em\u003e, 87(23), 9193\u0026ndash;9196. https://doi.org/10.1073/pnas.87.23.9193\u003c/li\u003e\n\u003cli\u003eSahan, S., Polat, K., Kodaz, H., \u0026amp; G\u0026uuml;nes, S. (2007). A new hybrid method based on fuzzy\u0026ndash;artificial immune system and k-nn algorithm for breast cancer diagnosis. \u003cem\u003eComputers in Biology and Medicine\u003c/em\u003e, 37(3), 415\u0026ndash;423. https://doi.org/10.1016/j.compbiomed.2006.05.003\u003c/li\u003e\n\u003cli\u003eShankarlingam G, Reddy KT. (2023) Predicting a Small Cap Company Stock Price using Python with Best Accuracy Rate: How the Data Science Working for Predictions and Accuracy Rate. Indian Journal of Science and Technology. 16(48): 4620- 4623. https://doi.org/10.17485/IJST/v16i48.2793 \u003c/li\u003e\n\u003cli\u003ethirupathi. A Streamlit-powered System for Simulating and Visualizing Qubit States, 11 September 2025, PREPRINT (Version 1) available at Research Square [https://doi.org/10.21203/rs.3.rs-7579530/v1]\u003c/li\u003e\n\u003cli\u003eKandadi, Thirupathi and Shankarlingam, G., Analysis Report of Traders Losing Money in Trading Platforms: A Sebi Report (October 26, 2024). ssrn, Available at SSRN: https://ssrn.com/abstract=5000105 or http://dx.doi.org/10.2139/ssrn.5000105 \u003c/li\u003e\n\u003cli\u003eKandadi, Thirupathi and Shankarlingam, G., DRAWBACKS OF LSTM ALGORITHM: A CASE STUDY (January 01, 2025). Available at SSRN: https://ssrn.com/abstract=5080605 or http://dx.doi.org/10.2139/ssrn.5080605\u003c/li\u003e\n\u003cli\u003eKandadi, Thirupathi, IMPLEMENTATION OF DICHOTOMISER 3 ALGORITHM WITH DECISION TREE FOR MAKING DECISIONS IN MACHINE LEARNING (April 07, 2025). Available at SSRN: https://ssrn.com/abstract=5207811 or http://dx.doi.org/10.2139/ssrn.5207811\u003c/li\u003e\n\u003cli\u003eKandadi Thirupathi Reddy. A FRAMEWORK TO PREDICTING STARTUP SUCCESS GROWTH WITH MULTIPLE ALGORITHMS. \u003cem\u003eTechRxiv.\u003c/em\u003e September 10, 2025. DOI: 10.36227/techrxiv.175751434.42657764/v1\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"chaitanya deemed to be university","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":"Breast Cancer Classification, Machine Learning, Diagnostic Accuracy, Malignant vs. Benign, Predictive Modeling, Medical Diagnosis, Data-driven Healthcare, Classification Metrics, Precision and Recall, Clinical Decision Support","lastPublishedDoi":"10.21203/rs.3.rs-7638704/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7638704/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eBreast cancer categorization is vital for prompt identification and strategic therapy planning. In this research, a machine learning model was created utilizing the Breast Cancer Wisconsin (Diagnostic) dataset to categorize tumors as malignant or benign. The model achieved an overall accuracy of 97.36%, demonstrating strong predictive performance.\u003c/p\u003e\u003cp\u003eFor the malignant class, the model attained a precision of 0.976, recall of 0.953, and F1-score of 0.965, indicating a strong accuracy in detecting cancerous instances. For the benign \u003cb\u003eclass\u003c/b\u003e, the model achieved a precision of 0.972, recall of 0.986, and F1-score of 0.979, confirming its effectiveness in correctly classifying non-cancerous cases. The macro average F1-score was 0.972, and the weighted average F1-score was 0.974, further emphasizing equitable performance in both categories.\u003c/p\u003e\u003cp\u003eThese findings indicate that the suggested classification method offers a reliable and precise diagnostic instrument, with possible uses in clinical decision support systems for breast cancer detection.\u003c/p\u003e","manuscriptTitle":"Machine Learning-based Breast Cancer Classification Using Logistic Regression and Random Forest","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-09-23 07:53:53","doi":"10.21203/rs.3.rs-7638704/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":"c7136d7e-eea2-41f7-84f1-3a8fd45d63f5","owner":[],"postedDate":"September 23rd, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":54911977,"name":"General Biochemistry"},{"id":54911978,"name":"Cancer Biology"},{"id":54911979,"name":"Artificial Intelligence and Machine Learning"},{"id":54911980,"name":"Cellular \u0026 Molecular Neuroscience"},{"id":54911981,"name":"Biophysics"}],"tags":[],"updatedAt":"2025-09-23T07:53:53+00:00","versionOfRecord":[],"versionCreatedAt":"2025-09-23 07:53:53","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7638704","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7638704","identity":"rs-7638704","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}