A Novel Healthcare Monitoring System using Optimal Hybrid DL, AR and IoT

A Novel Healthcare Monitoring System using Optimal Hybrid DL, AR and IoT | 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 A Novel Healthcare Monitoring System using Optimal Hybrid DL, AR and IoT G Ganesan, S Poonkuntran This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5244034/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 The growth of IoT in modern days helps devices to work productively and efficiently. However, the device needs human intervention for timely assistance. The proposed design consists of transmitter and receiver side, the transmitter side comprises of various sensors connected with the Raspberry Pi module and is connected to the camera and Wi-Fi module. The transmitted data are recognized and classified at the receiver side for any unstable healthcare condition of the employees or patients which could be connected and monitored over Augmented reality. Once the abnormal status is detected, an alert will be sent to the caretaker or physician to take necessary actions. The performance estimation is made in terms of various performance metrics and the comparison made with the existing system reveals that the proposed model is effective over other conventional schemes. IoT Augmented reality Expectation Maximization based clustering Tracking employee’s healthcare status Intelligent swarm dependent BAT optimization approach Hybrid Inception V3 and MobileNet V2 model. Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Figure 10 1 Introduction Recently, the meaning of health has got some new perceptions that focus on the physical and emotional status of person. As described by the World Health Organization [WHO], health represents not merely the lack of ability or disease however it is the condition of entire mental, physical, and social wellness focusing on the person’s physical and emotional well-being [Qadri et al., 2020]. Consequently, the system of healthcare continuously demands newer technologies that ranges from newer medicines to the complicated equipment for diagnosis with better plans on preventive-health [Malche et al., 2022]. On seeing those ideas, there is an enhancing tendency at the field of computer science and engineering that focus on developing newer tools which aid medical professionals in getting better help and diagnosis thus reducing the disease detection recognition cost. At this perspective, various physiological signals of patients [Shi et al., 2020]. Those systems might aid in conventional medicine to diagnose central and psychological purposes like recognizing mental physiological variations that takes place in organism states thereby giving interpreted feedback automatically. The Internet of Medical Things [IoMT] integrated with IoT technology and the medical applications might enable the realization of intelligent healthcare, precision medicine, and digital and intelligent age telemedicine [Patel & Bhatia, 2024]. Usual kinds of wearable sensors employed in IoMT usually covers photovoltaic sensors, resistive sensors, piezoelectric sensors, and capacitive sensors. These huge number of sensor nodes of IoMT makes usual replacement of battery a costly one and inconvenient at some conditions. Moreover, amount of information generated by sensors in the IoMT needs effective computation and processing. Moreover, there is a need for expansion of intelligent IoMT system having sustainable sources of power. Metaverse, interconnected network augmented spaces at which the users could interact & experience over the wearable devices, thus offering new opportunity for enhancing the quality of life for the patients/employees [S. Razdan et al., 2022]. The wearable sensors of IoMT system might be employed for project human movements, physiological signs, emotions, & environmental interaction of virtual world from the real world, thus transforming the experience of user in the intelligent healthcare, human-machine interaction, AR, and VR [virtual reality]. The technology of AR offers an auspicious choice for enriching the sensorial views & the interaction needs [Morris & Yeboah, 2023]. The AR system supplements the reality, that could not be embodied sufficiently in a label of object by means of superimposing virtual objects [computer-generated], like texts, graphics, and sounds over the real-world environment of the user. This could be imagined how the positioning of smart devices are facilitated by IoT devices & contents of AR [Joko, 2023] which might be aided from the visualization and interpretation of data with which the physician might interact with targeted health status and studies daily records over virtual contents. With the intention of using AR technology integrated with IoT/IoMT device for supporting visualization of data, in this manuscript, a novel design framework which integrated IoT with AR dependent framework termed AR-IoT is proposed. The main intention of this model is to design a framework to track employee’s healthcare status such as heart beat and brain health with the use of IoT devices. The health status monitoring is integrated with Augmented reality by employing DL models at which the recognition or tracking of normal and abnormal health status of employees are carried by processing the input data from sensors using Artificial Intelligence and Deep Learning models. From this, an alert message will be sent to the physician while abnormal tracking occurs which could be monitored and given necessary actions via Augmented reality technology. The residual sections of this manuscript are segregated as follows: section II is the related works offered related to this AR based IoT technology. The proposed design framework is narrated in section III. The analysis of performance made for proposed model is shown in section IV. The conclusions are provided in section V. 2 Related Work A short review of various conventional models related to AR technology integrated with IoT is projected in this section. The author of the work [Aditya & Dash, 2022] aims to implement IoT integration with the AR model to visualize and assess virtual data over real-time objects in a convenient manner. The step-wise process was given like how one could employ AR and IoT altogether which was not much narrated in review. The AR was employed for viewing data over real-world things. The mobile camera captures & records the video thus employing Vuforia SDK of any virtual data such as 3D shapes, and environments which could be overlaid in video. Therefore, the virtual content that comprises of 3D model of the machine together with sensor data could be shown in the real-world video. It was stated in [Gomes et al., 2020], that AR was the digital content overlay in the real-time world which thus plays a significant role in the product's lifecycle from their support to design, thereby permitting high flexibility. The interconnected system adoption and the utilization of IoT have driven the utilization of AI technology due to excess data that comes from varied sources will be unstructured. Various AI-based models were employed for decades thus aiming to make sense of unstructured data thus transforming them to suitable data. Hence, the AR, converging IoT, and AI thus makes the system autonomous enchantingly and a problem-solving one in various conditions. A study [Ghorbani, 2024] reported the outcome of the assistive intelligence system [IA], thus attained over the IoT integration, AR, & adaptive fuzzy decision-making models. The suggested model has 4 major mechanisms the location and stored heading data at the local fog layer, the AR device that interacts with the patients, a supervisory decision-making module for handling the environmental and direct interactions with patients, and the user interface for caregivers or family for monitoring the real-time situation of patient and sends reminders or alerts if needed. A new architecture on IAQ which was based on cognitive IoT [CIoT] with the use of AR was presented in the work [Sassi & Fourati, 2020]. The suggested model offers huge potential for integrating the IoT for the data of IAQ with the visualization of AR. This in turn allows instinctive communication between users & the devices of IoT which enhances the data visualization further that were composed of sensors of IoT that contribute effectively to improve the usability and experience of user. In the work [Kim et al., 2021], a comprehensive review was made of various proposed MI models which use AR and IoT. About 23 studies in total were analyzed and identified over the rigorous review protocol. The outcome of this MI system architecture model, their relationship among system components, and interaction modalities of input/output. Their open research challenges were discussed and presented for summarizing the results and thus identifies future development of research paths for MI developers and researchers. A system was proposed in the work [Naheem et al., 2023] at which the crucial data for physicians was superimposed at its real-time surroundings through the semi-transparent glasses housed in the AR headset. In this, sensors that were worn by the patients in hospitals collect information of real-worlds, thus processing them and sending them to the doctors' AR glasses through wireless for altering them regarding any such abnormality state of the patient. Depending on the present health status of the patient, the doctor could take necessary actions to improve the condition of the patient. A prototype platform for exergames was presented in the work [Koulouris et al., 2022] which integrated AR with IoT on the commodity of mobile devices to develop serious games regarding healthcare field. The major intention of this solution was to augment the use of gamification models for boosting the physical activities of the patient and for assisting them regarding the usual assessment of health and cognitive status over challenges & quests in a real and virtual world. A solution was thus validated in real-time scenarios and outcomes were interpreted to enhance the usability and performance of the prototype. The work [Elavarasi & Kavitha, 2023] presents, progresses, and thus authenticates AR & IoT-aided systems of healthcare to be employed by physicians. The suggested model was based on a smart city IoT-middleware structure offering an intuitive, standardized, and non-intrusive manner of delivering data on elder persons to their clinicians. This prototype was presented thus assessing the outcome which reveals the system's efficiency. The time of execution on average includes the detection of objects, thus communicating them with the server, and offering outcomes in the application. The suggested framework of AR/MR-IoT makes use of open-source and standard protocols like HTTPS, MQTT, or Node-RED [Blanco-Novoa et al., 2020]. The review made in [Sahu et al., 2024] sheds light on various components needed for improving the virtual and augmented reality-based technology which was followed by various devices employed in virtual and augmented-based systems. The usage of this model thereby signifies the neural flaws & presented approaches for employing those advancements thus offering an immersive experience for both educational and research purposes. Hence, the role of IoT with VR and AR soon is to diagnose and treat neurological disorders. AR is a promising technology as stated in [Aanjanadevi et al., 2022] which was the emerging key for unlocking the potential of IoT to the fullest. The challenges and potentials of integrating the IoT with AR were investigated here. After thorough process of searching, about 51 publications were selected and examined thoroughly for summarizing the recent developments more recently at this area. In the research identification, about 4 major clusters of potential advantages were found: sensor data visualization, interaction of human-object, diversity of application, and use case adoption of IoT. On the contrary, the limitations posed by AR and IoT integrations were thus clustered over organizational, technical, and ergonomic deliberations. The aim of this work [Siang et al., 2023] was to develop smart manufacturing plant integrated with AR and IoT interface so as to augment the controlling and monitoring process efficiency. At this work, an ASRS-based system was implemented and thus integrated with IoT and AR applications. The concluding prototype of this AR application allows user to monitor, control ASRS system in real-world. By the implementation of IoT and AR, efficiency of new ASRS system might augment hugely which enhances the control, monitor, and technical guidance support of system. Also, the interface of IoT might emerge for storing data or information related to the system, for instance, system status and storage information. 3 Proposed Work The proposed framework involves two stages the transmitter and receiver side. At the side of the transmitter, the sensors are connected with the Raspberry Pi module which is connected to the camera and Wi-Fi module. The output from this is transmitted to the receiver side where the input data are recognized and classified for any unstable healthcare condition of the employees or patients which could be connected and monitored over the Augmented reality. 3.1 Hardware sensors and data acquisition from the transmitter Presently, wearable sensors are regarded as a new tool for human activity recognition [HAR], placed in various parts of the human body. This is significant for knowing the precise body location of the wearable sensors and is the precise tool for attaching them to the human body. The location of the sensor on the body has an important impact on estimating the movements of the body and identifying the activities of the body, such that research is done in this field. The sensors could be placed and are visible and located typically on the sternum, waist, and belt. The wearable sensors over the placement of waist could monitor the movements of humans more precisely as this is nearer to the center of human body. The sensor numbers like the location of the sensor in HARS are needed. As per the research, the integration of thighs, chest, and ankles are embedded in the sensors and is more precise. Most often, smartphones incorporate entire kinds of sensors. The body temperature sensor, blood pressure, Heart rate sensor, Sp02 sensor, Blood Glucose sensor, and activity monitoring are placed in wearable device. The people employ wearable sensors for generating excess information regarding its movement, position, interaction, and location. The peripheral sensors might attain information regarding the environment of smart home like humidity, temperature, pressure, sound, light, and so on. They are not created for monitoring the activities in groups and thus discriminating them among residents’ actions or movements. Various sensors could perform several tasks on monitoring for measuring properties like position, movement, ECG, and temperature. The camera is likewise a traditional tool for carefully attaining information. Adequate 2-D data is thus provided from various viewpoints to extract 3-D movements on humans and thus the environments are pre-determined. This view of the area on fixed cameras is thus limited. Other limitation of cameras covers the fact that several people does not feel gratified that entire movements are under control. The wireless communication is made using a Wi-Fi module for transmitting the health care data to the IoT cloud module. Figure 1 a. Defines the sensors are connected to the camera and Wi-Fi module using a Raspberry Pi module and b. defines the AR with receiver output. 3.2 Input Pre-processing The input healthcare data of employees exhibits the subsequent artifacts: this is not complete and thus lacks fewer essential attributes or this must include collective data, as it is noisy and has some errors, this is not a consistent one and comprises similar codes or naming discrepancies. For carrying out such an analysis this is essential that a data must undergoes the pre-processing stage which covers cleaning, integration, transformation, reduction, and discretization. For carrying cleaning, the attributes that are missing in input data is thus filled thus identifying outliers. The noisy data is thus smoothened to rectify inconsistent data. The process of data transformation covers data normalization, aggregation, generalization, and attribute generation. In data reduction, attributes number and tuples are thus decreased together with the data number of attributes. So as to attain this, contours that were probable with specifications were identified for detecting abnormalities. 3.3 EM Based Clustering EM is the probabilistic and iterative approach that switches among the expectation [E] and maximization [m] stages consecutively. In E-phase, EM computes the likelihood function of expected values. In the M-phase, moreover, EM attains parameter estimation for maximizing the likelihood function. The attained parameters at the M-phase were employed at the subsequent E-phase. This process gets repeated till there is an occurrence of convergence that is it convergences to the parameter’s final values. For performing E-phase at the estimation of probabilities of the input data which belongs to cluster or model mixture, EM carried subsequent formula: $$\:{\varvec{p}\left(\varvec{m}\vee\:\varvec{n}\right)}^{\varvec{t}}=\frac{{\varvec{w}}_{\varvec{c}}^{\varvec{t}}\varvec{h}\left({\varvec{X}}_{\varvec{n}},{\varvec{\mu\:}}_{\varvec{m}}^{\varvec{t}},{\varvec{\sigma\:}}_{\varvec{m}}^{\varvec{t}}\right)}{{\sum\:}_{\varvec{k}=1}^{\varvec{M}}{\varvec{w}}_{\varvec{c}}^{\varvec{t}}\varvec{h}\left({\varvec{X}}_{\varvec{n}},{\varvec{\mu\:}}_{\varvec{m}}^{\varvec{t}},{\varvec{\sigma\:}}_{\varvec{m}}^{\varvec{t}}\right)}$$ 1 Here, h signifies the function of the probability density of the input \(\:{X}_{n}\) at dataset for m cluster having standard deviation \(\:{\sigma\:}_{m}^{t}\) , iteration t, and \(\:{\mu\:}_{m}^{t}\) mean. In this, the normal univariant, multivariate, or bivariate function of probability density might be utilized and this is according to the data dimensionality. Also, at the constraint \(\:{\sum\:}_{k=1}^{M}{w}_{k}^{t}=1\) , data allocation for the mixture models is then influenced through the weighting factor \(\:{w}^{t}\) . As per the M-phase performance, for the likelihood maximization function, EM should need the computation of subsequent each cluster estimation: $$\:{\mu\:}_{m}^{t+1}=\frac{{\sum\:}_{n=1}^{N}p{\left(m\vee\:n\right)}^{t}{x}_{n}}{{\sum\:}_{n=1}^{N}p{\left(m\vee\:n\right)}^{t}}$$ 2 $$\:{\sigma\:}_{m}^{t+1}=\sqrt{\frac{{\sum\:}_{n=1}^{N}p{\left(m\vee\:n\right)}^{t}{x}_{n}-{\mu\:}_{m}^{t+1}}{{\sum\:}_{n=1}^{N}p{\left(m\vee\:n\right)}^{t}}}$$ 3 $$\:{w}_{k}^{t+1}=\frac{{\sum\:}_{n=1}^{N}p{\left(m\vee\:n\right)}^{t}}{N}$$ 4 The above process gets repeated on executing M and E phases till the occurrence of convergence. 3.4 Feature extraction and optimal selection of features After the clustering process, the extraction of features is carried out to mine relevant and suitable features to help the classification process. It is necessary to employ an optimal feature selection module to select best optimal feature subset for which Intelligent swarm dependent BAT optimization process is carried. This proposed Intelligent swarm dependent BAT algorithm is a method of optimization working based on intelligence of bats. There is the emission of short pulses by bat for their hunting and movement process. The pattern of successive navigation depends on echo returned. The bats could predict the obstacle kind depending on the signal of echo. • The entire bats thus utilize the escalation process for predicting the distance & could differentiate between the obstacles and prey. • Bats could fly with the speed of v i having frequency of f min, with position \(\:{x}_{i}\) , \(\:{A}_{o}\) loudness & varying the wavelength λ to identify prey. • Bats could be able to update the frequency of their expelled pulse and emission of pulse rate which is updated \(\:r\) \(\:\in\:\left[\text{0,1}\right]\) . • Consider that the loudness varies from the maximum \(\:{A}_{o}\) value to the smaller \(\:{A}_{min}\) value \(\:.\) The velocities and positions of the bat were initialized randomly. The frequency of pulse emission is altered & the equation is shown by: $$\:{f}_{i}={f}_{min}+\left({f}_{max}-{f}_{min}\right)\beta\:$$ The velocity is thus altered & their equation is provided by: $$\:{v}_{i}^{t}={v}_{i}^{t-1}+\left({x}_{i}^{t}-{x}_{gbest}\right){f}_{i}$$ The bat position is thus altered and their equation is given by: $$\:{x}_{i}^{t}={x}_{i}^{t-1}+{v}_{i}^{t}$$ \(\:{f}_{min}\) – \(\:minimumfrequency\) \(\:{f}_{max}\) – \(\:maximumfrequency\) \(\:{v}_{i}^{t-1}\) denotes the preceding velocity & \(\:{x}_{gbest}\) denotes the optimal position. A newer solution is updated for its local position and is provided by the equation: \(\:{x}_{update}\) = \(\:{x}_{old}\) + \(\:\in\:{A}^{t}\) (8) \(\:\in\:\:\) signifies the random number among [-1,1]. The algorithm for this intelligent swarm-dependent BAT algorithm is provided below: Algorithm 1: Intelligent swarm dependent BAT optimization approach \(\:\text{I}\text{n}\text{i}\text{t}\text{i}\text{a}\text{l}\text{i}\text{z}\text{e}\:\text{b}\text{a}\text{t}\text{p}\text{o}\text{p}\text{u}\text{l}\text{a}\text{t}\text{i}\text{o}\text{n}\) as \(\:{\text{x}}_{\text{i}}\) , i = 1,2,……..n & frequencies \(\:{\text{f}}_{\text{i}},\text{s}\text{p}\text{e}\text{e}\text{d}{\text{v}}_{\text{i}}\) , \(\:\text{l}\text{o}\text{u}\text{d}\text{n}\text{e}\text{s}\text{s}\) \(\:{\text{A}}_{\text{i}}\) & rate of pulse \(\:{\text{r}}_{\text{i}}\) . \(\:\text{w}\text{h}\text{i}\text{l}\text{e}\:\left(\text{t}<\text{m}\text{a}\text{x}\text{i}\text{m}\text{u}\text{m}\:\text{i}\text{t}\text{e}\text{r}\text{a}\text{t}\text{i}\text{o}\text{n}\text{s}\right)\) \(\:\text{P}\text{r}\text{o}\text{d}\text{u}\text{c}\text{e}\:\text{n}\text{e}\text{w}\:\text{s}\text{o}\text{l}\text{u}\text{t}\text{i}\text{o}\text{n}\text{s}\:\text{b}\text{y}\:\text{a}\text{l}\text{t}\text{e}\text{r}\text{i}\text{n}\text{g}\:\text{t}\text{h}\text{e}\:\text{f}\text{r}\text{e}\text{q}\text{u}\text{e}\text{n}\text{c}\text{y}\) \(\:\text{M}\text{o}\text{d}\text{i}\text{f}\text{y}\:\text{t}\text{h}\text{e}\:\text{p}\text{o}\text{s}\text{i}\text{t}\text{i}\text{o}\text{n}\text{s}\wedge\:\text{v}\text{e}\text{l}\text{o}\text{c}\text{i}\text{t}\text{i}\text{e}\text{s}\) \(\:\text{i}\text{f}\) \(\:{\text{r}}_{\text{i}}\) \(\:\text{C}\text{h}\text{o}\text{o}\text{s}\text{e}\:\text{t}\text{h}\text{e}\:\text{o}\text{p}\text{t}\text{i}\text{m}\text{a}\text{l}\:\text{s}\text{o}\text{l}\text{u}\text{t}\text{i}\text{o}\text{n}\) \(\:\text{C}\text{r}\text{e}\text{a}\text{t}\text{e}\:\text{a}\:\text{l}\text{o}\text{c}\text{a}\text{l}\:\text{s}\text{o}\text{l}\text{u}\text{t}\text{i}\text{o}\text{n}\:\text{a}\text{r}\text{o}\text{u}\text{n}\text{d}\:\text{t}\text{h}\text{e}\:\text{o}\text{p}\text{t}\text{i}\text{m}\text{a}\text{l}\) \(\:\text{e}\text{n}\text{d}\text{i}\text{f}\) \(\:\text{C}\text{r}\text{e}\text{a}\text{t}\text{e}\:\text{a}\:\text{o}\text{r}\text{i}\text{g}\text{i}\text{n}\text{a}\text{l}\:\text{s}\text{o}\text{l}\text{u}\text{t}\text{i}\text{o}\text{n}\:\text{b}\text{y}\:\text{r}\text{a}\text{n}\text{d}\text{o}\text{m}\) \(\:\text{i}\text{f}\) [random < \(\:{\text{A}}_{\text{i}}\) ] and [f[ \(\:{\text{x}}_{\text{i}}<\) f[ \(\:{\text{x}}_{\text{g}\text{b}\text{e}\text{s}\text{t}}]\) \(\:\text{i}\text{d}\text{e}\text{n}\text{t}\text{i}\text{f}\text{y}\:\text{t}\text{h}\text{e}\:\text{o}\text{r}\text{i}\text{g}\text{i}\text{n}\text{a}\text{l}\:\text{s}\text{o}\text{l}\text{u}\text{t}\text{i}\text{o}\text{n}\) \(\:\text{U}\text{p}\text{d}\text{a}\text{t}\text{e}\:\text{b}\text{y}\:\text{i}\text{n}\text{c}\text{r}\text{e}\text{a}\text{s}\text{i}\text{n}\text{g}\) \(\:{\text{r}}_{\text{i}}\) & \(\:\text{d}\text{e}\text{c}\text{r}\text{e}\text{a}\text{s}\text{e}\) \(\:{\text{A}}_{\text{i}}\) \(\:\text{e}\text{n}\text{d}\text{i}\text{f}\) \(\:\text{F}\text{i}\text{n}\text{d}\:\text{t}\text{h}\text{e}\:\text{c}\text{u}\text{r}\text{r}\text{e}\text{n}\text{t}\:\text{b}\text{e}\text{s}\text{t}\wedge\:\text{r}\text{a}\text{n}\text{k}\:\text{t}\text{h}\text{e}\:\text{b}\text{a}\text{t}\text{s}\) \(\:\text{e}\text{n}\text{d}\text{w}\text{h}\text{i}\text{l}\text{e}\) Consequently, the optimal range of features are extracted and selected utilizing this proposed optimization approach. 3.5 Hybrid Inception v3 and MobileNet v2 Classifier model The presented hybrid MobileNet v2 and Inception v3 model is utilized to predict the classification process. Figure 2 . Defines the MobileNet v2 structure consists of the convolutional layers at each of them is then followed by the batch normalization layer and ReLU non-linear activation function which excludes the output layer. The convolutional layers are expensive in the MobileNet V2 which is factorized as the lightweight convolutional blocks which are separable. The input block is filtered at each block over the depth-wise convolutional layer having size [3x3]. The output channels offered are thus projected over the pointwise convolutional layer having size [1x1]. The MobileNet v2 blocks are built usually with residual connectivity that helps converge network weight. The depth wise utilization with 3x3 removeable convolutions & the resolutions modification over layers infers on decreasing the of computation complexity that corresponds to the standardized form of convolutions despite minor variation in accuracy. For the image classification, Inception V3 is likewise employed which is extended typically from the network of GoogleNet expressed in Figure. 3. This relies on offering several smaller convolutions having similar level instead of employing huge convolutional ranges. The first layer of network consists of three convolutions and one max-pooling layer. The final one comprises of channels that are provided and are merged non-linearly. Therefore, the network parameter reduction is carried which implies on the training or testing accelerations stages. Likewise, the features are extracted effectively which increases the classification accuracy. This model consists of dense layers to increase the network depth that reduces complexity of computation once more. This technique provides higher rate of accuracy than other traditional techniques like ResNet, AlexNet, and GoogLeNet. Due to its flexibility and accuracy, this technique offers higher performance. The suggested model employs score-dependent functions for estimating score values & creates prediction effectively. The computation of score value is made by means of following equation: $$\:score={\sum\:}_{i=1}^{M}{\sum\:}_{j=1}^{N}\left({A}_{1}\left(i,j\right)+{A}_{2}\left(i,j\right)\right)/2$$ 9 In this, A denotes the score generated by classifier model presented. N signifies number of classes & M denotes test sets. On employing above equation, new score having M×N size is achieved. After that, using this new score value, class labels are determined as per the maximum value of each line [M]. The algorithm for this is shown below: Algorithm 2: Proposed Hybrid classification technique \(\:\text{I}\text{n}\text{p}\text{u}\text{t}:\text{s}\text{c}\text{o}\text{r}\text{e}1,\text{s}\text{c}\text{o}\text{r}\text{e}2,\text{Y}\text{T}\text{e}\text{s}\text{t}\) \(\:\text{O}\text{u}\text{t}\text{p}\text{u}\text{t}:\text{a}\text{c}\text{c}\text{u}\text{r}\text{a}\text{c}\text{y}\) \(\:\text{S}\text{t}\text{e}\text{p}1:\:\text{f}\text{o}\text{r}\:\text{i}=1\:\text{t}\text{o}\:\text{M}\) \(\:\text{S}\text{t}\text{e}\text{p}2:\) \(\:\text{f}\text{o}\text{r}\:\text{j}=1\:\text{t}\text{o}\:\text{N}\) \(\:\text{S}\text{t}\text{e}\text{p}3\) : \(\:{\text{n}\text{e}\text{w}\text{e}\text{r}}_{\text{s}\text{c}\text{o}\text{r}\text{e}}\left(\text{i},\text{j}\right)=\left(\text{s}\text{c}\text{o}\text{r}\text{e}1\left(\text{i},\text{j}\right)+\text{s}\text{c}\text{o}\text{r}\text{e}2\left(\text{i},\text{j}\right)\right)/2\) \(\:\text{S}\text{t}\text{e}\text{p}4:\) \(\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\text{e}\text{n}\text{d}\text{f}\text{o}\text{r}\:\text{j}\) \(\:\text{S}\text{t}\text{e}\text{p}5:\) \(\:\text{e}\text{n}\text{d}\text{f}\text{o}\text{r}\:\text{i}\) \(\:\text{S}\text{t}\text{e}\text{p}6:\:\text{f}\text{o}\text{r}\:\text{k}=1\text{t}\text{o}\:\text{K}\) \(\:\text{S}\text{t}\text{e}\text{p}7:\) \(\:\text{i}\text{f}\:\text{s}\text{c}\text{o}\text{r}\text{e}\left(\text{k},1\right)\ge\:\text{s}\text{c}\text{o}\text{r}\text{e}\left(\text{i},2\right)\) \(\:\text{S}\text{t}\text{e}\text{p}8:\) \(\:\text{Y}\text{P}\text{r}\text{e}\text{d}\left(\text{k}\right)=1;\) \(\:\text{S}\text{t}\text{e}\text{p}9:\) \(\:\text{e}\text{l}\text{s}\text{e}\) \(\:\text{S}\text{t}\text{e}\text{p}11:\) \(\:\text{Y}\text{P}\text{r}\text{e}\text{d}\left(\text{k}\right)=1;\) \(\:\text{S}\text{t}\text{e}\text{p}12:\:\text{a}\text{c}\text{c}\text{u}\text{r}\text{a}\text{c}\text{y}=\text{m}\text{e}\text{a}\text{n}\left(\text{Y}\text{p}\text{r}\text{e}\text{d}=\text{Y}\text{T}\text{e}\text{s}\text{t}\right);\) S \(\:\text{t}\text{e}\text{p}13:\) \(\:\text{e}\text{n}\text{d}\text{f}\text{o}\text{r}\:\text{k}\) The Hybrid proposed classifier model predicts the employee’s healthcare abnormality and offers classification outcome with high range of accuracy. This performance is estimated and is provided in subsequent section. 4 The Performance Analysis The perpetuation of experiments is evaluated for the suggested model with varied instance numbers that range from 10% to 100% and are compared with traditional models like LSTM, FIS integrated with LSTM [FLSTM], and Bi-LSTM [Nancy et al., 2022]. The comparison is made to show the effectiveness of the proposed model over other existing models related to the most widely used approach LSTM. Table 1: Comparative analysis of accuracy Data [%] Bi-LSTM 10 94.04 94.56 94.86 96.87 20 94.35 95.59 96.02 97.98 30 94.52 96.21 96.76 98.54 40 94.68 96.68 97.25 98.90 50 94.75 97.01 97.65 98.95 60 94.86 97.31 97.96 98.99 70 94.89 97.51 98.24 99.04 80 94.95 97.70 98.48 99.65 90 95.00 97.92 98.67 99.73 100 95.07 98.04 98.86 99.89 Table 1 shows the comparative analysis of accuracy for varied number of data for both proposed and existing models. The analysis estimated reveals that the proposed model is better and is enhanced than other traditional models. The graphical representation of this is shown in figure 4. Table 2 signifies the comparative examination of precision for the varied number of data for both proposed and existing models. The analysis estimated reveals that the suggested technique’s precision is augmented and is better than other traditional models. The graphical representation of this is shown in figure 5. Table 2: Comparative analysis of precision Data [%] Bi-LSTM 10 94.00 94.60 94.81 96.42 20 94.37 95.55 96.00 97.28 30 94.52 96.21 96.74 98.33 40 94.67 96.69 97.22 98.49 50 94.72 97.04 97.69 98.50 60 94.86 97.30 97.95 98.86 70 94.85 97.50 98.21 99.01 80 94.99 97.71 98.50 99.35 90 95.00 97.95 98.63 99.73 100 95.07 98.03 98.90 99.88 Table 3 represents the comparative analysis of recall for varied number of data for both proposed and existing models. The analysis estimated reveals that the proposed model recall is better and is enhanced than other traditional models. The graphical representation of this is shown in figure 6. Table 3: Comparative analysis of recall Data [%] LSTM FLSTM Bi-LSTM Proposed 10 94.08 94.51 94.90 95.11 20 94.32 95.63 96.04 96.27 30 94.52 96.21 96.78 96.38 40 94.69 96.68 97.29 97.40 50 94.79 97.00 97.61 97.88 60 94.87 97.32 97.97 97.99 70 94.92 97.53 98.27 98.67 80 94.91 97.68 98.45 98.70 90 95.00 97.89 98.72 99.63 100 95.06 98.03 98.81 99.91 Table 4 signifies the comparative analysis of specificity for varied data numbers for both proposed and existing models. The estimated analysis exposes that the proposed model specificity is better and is enhanced than other traditional models. The graphical depiction of this is revealed in Figure 7. Table 4: Comparative analysis of specificity Data [%] LSTM FLSTM Bi-LSTM Proposed 10 94.00 94.60 94.81 95.77 20 94.37 95.56 96.01 96.43 30 94.52 96.21 96.74 96.89 40 94.67 96.69 97.22 97.51 50 94.72 97.04 97.68 97.74 60 94.86 97.30 97.95 97.99 70 94.85 97.50 98.21 98.61 80 94.99 97.71 98.50 98.94 90 95.00 97.95 98.63 99.36 100 95.07 98.03 98.90 99.88 Table 5 indicates the comparative study of the F1-score for varied data numbers for both proposed and existing models. The analysis estimated exposes that the suggested scheme F1-score is improved and enhanced than other traditional models. The graphical illustration of this is provided in figure 8. Table 5. comparative analysis of F1-score Data [%] LSTM FLSTM Bi-LSTM Proposed 10 94.04 94.56 94.85 95.88 20 94.35 95.59 96.02 96.47 30 94.52 96.21 96.76 96.90 40 94.68 96.68 97.25 97.52 50 94.75 97.02 97.65 97.84 60 94.86 97.31 97.96 98.09 70 94.88 97.51 98.24 98.75 80 94.95 97.70 98.48 98.79 90 95.00 97.92 98.67 99.03 100 95.07 98.03 98.86 99.15 Table 6 represents the overall comparative analysis of performance for both proposed and existing models in terms of precision,. The estimated analysis shows that the proposed model outcome is better and is improved than other traditional models. The graphical illustration of this is provided in figure 9. Table 6: Overall comparison of performance Performance Metrics LSTM FLSTM Bi-LSTM Proposed Accuracy 95.07 98.04 98.86 99.89 Specificity 95.07 98.03 98.90 99.88 Recall 95.06 98.04 98.81 99.91 Precision 95.07 98.03 98.90 99.88 F1-score 95.07 98.03 98.86 99.15 Table 7 signifies the overall comparative performance analysis for both proposed and various existing models in terms of accuracy. The estimated analysis displays that the suggested scheme’s accuracy is higher & is augmented by comparing conventional models. https://www.kaggle.com/code/venkat1949/topic-modeling-based-biobert-qa-system. Table 7: Accuracy Comparison with existing models Models Accuracy Ensemble classifiers [Latha & Jeeva, 2019] 85.4 Hybrid RF and linear model [Mohan et al., 2019] 88.70 Type-2 fuzzy logic [Long et al., 2015] 86.0 Fuzzy analytic hierarchy & ANN [Long et al., 2015] 91.0 Statistical model & DNN [L. Ali et al., 2019] 91.57 Adaptive neuro-fuzzy [Paul et al., 2018] 92.3 Relief feature selection & DT [28] 92.8 [Ahmed et al., 2020] 94.15 CNN [Kishore & Jayanthi, 2018] 97.0 Fuzzy rules & DNN [Mehmood et al., 2021] 96.5 [Van Pham et al., 2018] 94.78 Sequential forward selection & RF [Dileep et al., 2023] 98.0 Kernel RF [Jabeen et al., 2019] 98.0 DCNN [Muzammal et al., 2020] 98.2 Ensemble DL [36] 98.5 Linear SVC & DNN [37] 98.56 Fuzzy information system & Bi-LSTM [21] 98.86 Proposed 99.89 Figure 10 signifies the graphical illustration of the overall comparative analysis of performance for both proposed and various existing schemes in terms of accuracy. The estimated analysis displays that the suggested model accuracy is better and is augmented over various existing methodologies. 5 Conclusion An Augmented reality-based IoT model for monitoring or tracking employees’ healthcare status was proposed in this work. The proposed design consists of transmitter and receiver side at which the transmitter side comprises of various sensors connected with Raspberry Pi module and is connected to camera and Wi-Fi module. The output from this is transmitted to the receiver side to process input cloud data for detecting abnormal health status using pre-processing, Expectation Maximization [EM] based clustering, and an Intelligent swarm-dependent BAT optimization approach to select the best features. and Hybrid Inception V3 and MobileNet V2 model for classification. In case, the abnormal status is detected, an alert will be sent to the caretaker or physician to take necessary actions. the performance estimation is made in terms of various performance metrics and the comparison made with the existing system reveals that the proposed model was improved over other existing models. 6 Future Work The challenges faced in such approaches are the security breaches like eavesdropping, jamming, denial of service [DOS] attacks, and spoofing attacks can be identified and control using proposed enhanced security with comprehensive future directions by creating IoB [Internet of Things Behaviour] to fight against attacks. The augmented approach gives a simulation approach to track the breaches that occurs while communicating with the devices. Declarations Funding “The authors declare that no funds, grants, or other support were received during the preparation of this manuscript.” Competing Interests “The authors have no relevant financial or non-financial interests to disclose.” Author Contributions “All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by Mr.G.Ganesan and Dr.S.Poonkuntran. The first draft of the manuscript was written by Mr.G.Ganesan and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.” Data Availability “The datasets generated during and/or analysed during the current study are not publicly available due to this work will be submitted for the author’s ph.d thesis but are available from the corresponding author on reasonable request..” References [Aanjanadevi et al., 2022] Aanjanadevi, S., Palanisamy, V., Aanjankumar, S., Poonkuntran, S., & Karthikeyan, P. [2022]. Independent automobile intelligent motion controller and redirection, using a deep learning system. In Object Detection with Deep Learning Models [pp. 165–178]. Chapman and Hall/CRC. [Aditya & Dash, 2022] Aditya, V. K., & Dash, A. K. [2022]. February],” Real-life Applications of integration of Augmented Reality and Internet of Things. In Second International Conference on Artificial Intelligence and Smart Energy [ICAIS] [pp. 1268–1273]. IEEE. [Ahmed et al., 2020] Ahmed, H., Younis, E. M., Hendawi, A., & Ali, A. A. [2020]. Heart disease identification from patients’ social posts, machine learning solution on Spark”. Future Generation Computer Systems , 111 , 714–722. [F. Ali et al., 2020] Ali, F., El-Sappagh, S., Islam, S. M. R., Kwak, D., Ali, A., Imran, M., & Kwak, K.-S. [2020]. A smart healthcare monitoring system for heart disease prediction based on ensemble deep learning and feature fusion. An International Journal on Information Fusion , 63 , 208–222. https://doi.org/10.1016/j.inffus.2020.06.008. [L. Ali et al., 2019] Ali, L., Rahman, A., Khan, A., Zhou, M., Javeed, A., & Khan, J. A. [2019]. An automated diagnostic system for heart disease prediction based on $\chi^2$ statistical model and optimally configured deep neural network. IEEE Access: Practical Innovations, Open Solutions , 7 , 34938–34945. https://doi.org/10.1109/access.2019.2904800. [Blanco-Novoa et al., 2020] Blanco-Novoa, Ó., Fraga-Lamas, P., Vilar-Montesinos, A., & Fernández-Caramés, M. [2020]. Creating the internet of augmented things: An open-source framework to make IoT devices and augmented and mixed reality systems talk to each other”. Sensors , 20 [11]. [Dileep et al., 2023] Dileep, P., Rao, K. N., Bodapati, P., Gokuruboyina, S., Peddi, R., Grover, A., & Sheetal, A. [2023]. An automatic heart disease prediction using cluster-based bi-directional LSTM [C-BiLSTM] algorithm. Neural Computing and Applications, 35[10], 7253–7266. [Elavarasi & Kavitha, 2023] Elavarasi, D., & Kavitha, R. [2023]. December]. Navigating Heart Health with an Elephantine Approach in Clinical Decision Support Systems. In 2nd International Conference on Automation, Computing and Renewable Systems [ICACRS] [pp. 1416–1423]. IEEE. [Ghorbani, 2024] Ghorbani, F. [2024]. An Intelligent Assistive System Based on Augmented Reality and Internet of Things for Patients with Alzheimer’s Disease. [Gomes et al., 2020] Gomes, P., Magaia, N., & Neves, N. [2020]. Industrial and artificial Internet of Things with augmented reality. In Convergence of Artificial Intelligence and the Internet of Things [pp. 323–346]. [Jabeen et al., 2019]Jabeen, F., Maqsood, M., Ghazanfar, M. A., Aadil, F., Khan, S., Khan, M. F., & Mehmood, I. [2019]. An IoT based efficient hybrid recommender system for cardiovascular disease”. Peer-to-Peer Networking and Applications, 12, 1263–1276. [Joko, 2023] Joko, S. [2023]. Development of AR Android Application for Disability and Elderly-Friendly Smart Home Control”. International Journal of Information Technology and Computer Engineering [IJITC], 4[02], 32–42. [Khan, 2020] Khan, M. A. [2020]. An IoT framework for heart disease prediction based on MDCNN classifier. IEEE Access: Practical Innovations, Open Solutions, 8, 34717–34727. https://doi.org/10.1109/access.2020.2974687. [Kim et al., 2021] Kim, J. C., Laine, T. H., & Ahlund, C. [2021]. Multimodal interaction systems based on internet of things and augmented reality: A systematic literature review”. Applied Sciences, 11[4]. [Kishore & Jayanthi, 2018] Kishore, A. H., & Jayanthi, V. E. [2018]. Neuro-fuzzy based medical decision support system for coronary artery disease diagnosis and risk level prediction”. Journal of Computational and Theoretical Nanoscience, 15[3], 1027–1037. [Koulouris et al., 2022] Koulouris, D., Menychtas, A., & Maglogiannis, I. [2022]. An IoT-enabled platform for the assessment of physical and mental activities utilizing augmented reality exergaming. Sensors [Basel, Switzerland], 22[9], 3181.https://doi.org/10.3390/s22093181. [Latha & Jeeva, 2019] Latha, C. B. C., & Jeeva, S. C. [2019]. Improving the accuracy of prediction of heart disease risk based on ensemble classification techniques”. Informatics in Medicine Unlocked, 16. [Long et al., 2015] Long, N. C., Meesad, P., & Unger, H. [2015]. A highly accurate firefly based algorithm for heart disease prediction”. Expert Systems with Applications , 42 [21], 8221–8231. [Malche et al., 2022] Malche, T., Tharewal, S., Tiwari, P. K., Jabarulla, M. Y., Alnuaim, A. A., Hatamleh, W. A., & Ullah, M. A. [2022] AIoT- based patient activity tracking system for remote patient monitoring. Artificial Intelligence of Things. [Mehmood et al., 2021] Mehmood, A., Iqbal, M., Mehmood, Z., Irtaza, A., Nawaz, M., Nazir, T., & Masood, M. [2021]. Prediction of heart disease using deep convolutional neural networks”. Arabian Journal for Science and Engineering , 46 [4], 3409–3422. [Mohan et al., 2019] Mohan, S., Thirumalai, C., & Srivastava, G. [2019]. Effective heart disease prediction using hybrid machine learning techniques. IEEE Access: Practical Innovations, Open Solutions, 7, 81542–81554. https://doi.org/10.1109/access.2019.2923707. [Morris & Yeboah, 2023] Morris, A., & Yeboah, G. [2023]. Convergence of IoT and Augmented Reality”. In Springer Handbook of Augmented Reality [pp. 831–851]. Springer International Publishing. [Muzammal et al., 2020] Muzammal, M., Talat, R., Sodhro, A. H., & Pirbhulal, S. [2020]. A multi-sensor data fusion enabled ensemble approach for medical data from body sensor networks. An International Journal on Information Fusion, 53, 155–164. https://doi.org/10.1016/j.inffus.2019.06.021. [Naheem et al., 2023] Naheem, M. R., Aleema, A., Fathima, K., Imran, K., & Fathima, K. [2023] Augmented Reality with Internet of Things. 5 , 2582–5208. [Nancy et al., 2022] Nancy, A. A., Ravindran, D., Vincent, P. D., Srinivasan, K., & Gutierrez Reina, D. [2022]. IoT-cloud-based smart healthcare monitoring system for heart disease prediction via deep learning. Electronics, 11[15]. [Patel & Bhatia, 2024] Patel, B., & Bhatia, J. [2024]. A comprehensive review of internet of things and cutting-edge technologies empowering smart farming. Current Science, 2. [Paul et al., 2018] Paul, A. K., Shill, P. C., Rabin, M. R. I., & Murase, K. [2018]. Adaptive weighted fuzzy rule-based system for the risk level assessment of heart disease, Applied Intelligence, 48, 1739–1756. [Qadri et al., 2020] Qadri, Y. A., Nauman, A., Zikria, Y. B., Vasilakos, A. V., & Kim, S. W. [2020]. The future of healthcare internet of things: a survey of emerging technologies. IEEE Communications Surveys & Tutorials, 22[2], 1121–1167. [S. Razdan et al., 2022] Razdan, S., Sharma, S., Manickam, P., Mariappan, S. A., Murugesan, S. M., Hansda, S., Kaushik, A., Shinde, R., & Thipperudraswamy, S. P. [2022]. Artificial intelligence [AI] and internet of medical things [IoMT] assisted biomedical systems for intelligent healthcare. Biosensors , 12 [8]. [Sahshanu Razdan & Sharma, 2022] Razdan, Sahshanu, & Sharma, S. [2022]. Internet of medical things [IoMT]: Overview, emerging technologies, and case studies. IETE Technical Review , 39 [4], 775–788. https://doi.org/10.1080/02564602.2021.1927863. [Sahu et al., 2024] Sahu, M., Gupta, R., Ambasta, R. K., & Kumar, P. [2024]. IoT-driven Augmented Reality and Virtual Reality Systems in Neurological Sciences, Internet of Things, 101098. [Samuel et al., 2017] Samuel, O. W., Asogbon, G. M., Sangaiah, A. K., Fang, P., & Li, G. [2017]. An integrated decision support system based on ANN and Fuzzy AHP for heart failure risk prediction. Expert Systems with Applications , 68 , 163–172. [Sassi & Fourati, 2020] Sassi, H., & Fourati, M. S. [2020]. Architecture for visualizing indoor air quality data with augmented reality based cognitive Internet of Things. In Advanced Information Networking and Applications: Proceedings of the 34th International Conference on Advanced Information Networking and Applications [AINA-2020] [pp. 405–418]. Springer International Publishing. [Shi et al., 2020] Shi, Q., Dong, B., He, T., Sun, Z., Zhu, J., Zhang, Z., & Lee, C. [2020]. Progress in wearable electronics/photonics—Moving toward the era of artificial intelligence and internet of things. InfoMat, 2[6], 1131–1162. https://doi.org/10.1002/inf2.12122. [Siang et al., 2023] Siang, H. T., Gobee, S., & Durairajah, V. [2023]. Development of augmented reality control and monitoring for smart manufacturing plant with IoT interface. In AIP Conference Proceedings [Vol. 2788]. AIP Publishing. [Van Pham et al., 2018] Van Pham, H., Son, L. H., & Tuan, L. M. [2018]. A proposal of expert system using deep learning neural networks and fuzzy rules for diagnosing heart disease. In Frontiers in Intelligent Computing: Theory and Applications: Proceedings of the 7th International Conference on FICTA [Vol. 1, pp. 189–198]. Springer. [Zhang et al., 2021] Zhang, D., Chen, Y., Chen, Y., Ye, S., Cai, W., Jiang, J., & Chen, M. [2021]. Heart disease prediction based on the embedded feature selection method and deep neural network”. Journal of Healthcare Engineering, 2021, 1–9. 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-5244034","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":382662580,"identity":"36971ae4-d1cf-4c32-aa0b-a84bebbffc60","order_by":0,"name":"G Ganesan","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABEElEQVRIie2NMUvDQBTHXwjY5YLrQUvyFe4INAQERz+DW0PAKbgLGR4IdSl2VfwSjh3fcZApxVWwS5ZOGewicVB61YIiXnUUvB88+N/j/e4P4HD8aXoI5CGZ5OPmzX5WGAGpN8X7rcJHAJ8VK9HRXDVduYDouJV6NVuESU8hdCUMEosi69M8ZtUS5EMhSNXLOJ1k6E0qYKmlTGIx7MOeBnlzMiI11tktZQgBAhNkUaZt8ty9GuW62ip3DXovO5SIF0MIxtoEn96V+wz9XS2Ct3E/uNRMsJxoXus4vWpQDypub5kWctU96TC6UOePZzMdJvu5atry4NDast1//CnMmMy/v9+04NcgrLcOh8PxX1kD6XVpNNoHtJoAAAAASUVORK5CYII=","orcid":"","institution":"VIT Bhopal University","correspondingAuthor":true,"prefix":"","firstName":"G","middleName":"","lastName":"Ganesan","suffix":""},{"id":382662583,"identity":"75f00a46-dca6-4b40-a143-7617a5dcc0d4","order_by":1,"name":"S Poonkuntran","email":"","orcid":"","institution":"VIT Bhopal University","correspondingAuthor":false,"prefix":"","firstName":"S","middleName":"","lastName":"Poonkuntran","suffix":""}],"badges":[],"createdAt":"2024-10-11 07:08:24","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5244034/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5244034/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":71612125,"identity":"77927cef-f396-426b-86d5-342bd050393b","added_by":"auto","created_at":"2024-12-17 06:58:24","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":32801,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003e[a] transmitter side [b] receiver side\u003c/em\u003e\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-5244034/v1/dd6584f65b65869764bbe677.png"},{"id":71608396,"identity":"b14bc019-c936-47b4-aca5-f7fad599aa44","added_by":"auto","created_at":"2024-12-17 06:34:24","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":25169,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eMobileNet v2 architecture model\u003c/em\u003e\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-5244034/v1/aae7250946f37fde098c95b2.png"},{"id":71610155,"identity":"3c460984-59fd-4dfb-8b19-3c785959c314","added_by":"auto","created_at":"2024-12-17 06:42:24","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":87273,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eInception v3 architecture model\u003c/em\u003e\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-5244034/v1/c8da0d6796faf635abfc6bbe.png"},{"id":71608398,"identity":"92702ba6-a13e-42b5-952f-1753085ebd49","added_by":"auto","created_at":"2024-12-17 06:34:24","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":13286,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eAccuracy-comparative analysis\u003c/em\u003e\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-5244034/v1/66f83e68add83a2f03b904f9.png"},{"id":71610646,"identity":"6b7440ab-a7d4-416d-99e7-b208a87d8c50","added_by":"auto","created_at":"2024-12-17 06:50:24","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":12909,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eComparative analysis of precision\u003c/em\u003e\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-5244034/v1/f419cdcb59c81dfbafa96135.png"},{"id":71610152,"identity":"723ea483-69f2-421d-9d3d-44012be778f3","added_by":"auto","created_at":"2024-12-17 06:42:24","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":13962,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eComparative analysis of recall\u003c/em\u003e\u003c/p\u003e","description":"","filename":"6.png","url":"https://assets-eu.researchsquare.com/files/rs-5244034/v1/322708097b4eef3771373f6f.png"},{"id":71610156,"identity":"04af9c0c-41c7-4a09-98f8-2c7587bab0e6","added_by":"auto","created_at":"2024-12-17 06:42:24","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":14604,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003especificity-comparative analysis\u003c/em\u003e\u003c/p\u003e","description":"","filename":"7.png","url":"https://assets-eu.researchsquare.com/files/rs-5244034/v1/437ff166cc761ee79ebd97d7.png"},{"id":71608404,"identity":"d2719b48-8f3f-49a5-9586-cf9b8c0404a8","added_by":"auto","created_at":"2024-12-17 06:34:24","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":11966,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003ecomparative analysis of F1-score\u003c/em\u003e\u003c/p\u003e","description":"","filename":"8.png","url":"https://assets-eu.researchsquare.com/files/rs-5244034/v1/c337d4e681ea765a146a203e.png"},{"id":71610648,"identity":"ef0bde44-6f4c-43f2-af0c-4e88a690e5fe","added_by":"auto","created_at":"2024-12-17 06:50:24","extension":"png","order_by":9,"title":"Figure 9","display":"","copyAsset":false,"role":"figure","size":13380,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eOverall comparison of performance\u003c/em\u003e\u003c/p\u003e","description":"","filename":"9.png","url":"https://assets-eu.researchsquare.com/files/rs-5244034/v1/71fa08cbaaf41b92de800052.png"},{"id":71608401,"identity":"f5304d2c-6b7c-462f-822d-b60c188ea5d1","added_by":"auto","created_at":"2024-12-17 06:34:24","extension":"png","order_by":10,"title":"Figure 10","display":"","copyAsset":false,"role":"figure","size":44456,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eAccuracy comparison with existing models\u003c/em\u003e\u003c/p\u003e","description":"","filename":"10.png","url":"https://assets-eu.researchsquare.com/files/rs-5244034/v1/16b7c4d40e97fef6fd47a2c1.png"},{"id":76026217,"identity":"fab6f900-872f-4e36-b803-b5406f902c5e","added_by":"auto","created_at":"2025-02-11 14:32:04","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1083980,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5244034/v1/ec10b160-6b10-4f3b-9388-da3988eb0e9e.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"A Novel Healthcare Monitoring System using Optimal Hybrid DL, AR and IoT","fulltext":[{"header":"1 Introduction","content":"\u003cp\u003eRecently, the meaning of health has got some new perceptions that focus on the physical and emotional status of person. As described by the World Health Organization [WHO], health represents not merely the lack of ability or disease however it is the condition of entire mental, physical, and social wellness focusing on the person\u0026rsquo;s physical and emotional well-being [Qadri et al., 2020]. Consequently, the system of healthcare continuously demands newer technologies that ranges from newer medicines to the complicated equipment for diagnosis with better plans on preventive-health [Malche et al., 2022]. On seeing those ideas, there is an enhancing tendency at the field of computer science and engineering that focus on developing newer tools which aid medical professionals in getting better help and diagnosis thus reducing the disease detection recognition cost. At this perspective, various physiological signals of patients [Shi et al., 2020]. Those systems might aid in conventional medicine to diagnose central and psychological purposes like recognizing mental physiological variations that takes place in organism states thereby giving interpreted feedback automatically.\u003c/p\u003e \u003cp\u003eThe Internet of Medical Things [IoMT] integrated with IoT technology and the medical applications might enable the realization of intelligent healthcare, precision medicine, and digital and intelligent age telemedicine [Patel \u0026amp; Bhatia, 2024]. Usual kinds of wearable sensors employed in IoMT usually covers photovoltaic sensors, resistive sensors, piezoelectric sensors, and capacitive sensors. These huge number of sensor nodes of IoMT makes usual replacement of battery a costly one and inconvenient at some conditions. Moreover, amount of information generated by sensors in the IoMT needs effective computation and processing. Moreover, there is a need for expansion of intelligent IoMT system having sustainable sources of power. Metaverse, interconnected network augmented spaces at which the users could interact \u0026amp; experience over the wearable devices, thus offering new opportunity for enhancing the quality of life for the patients/employees [S. Razdan et al., 2022]. The wearable sensors of IoMT system might be employed for project human movements, physiological signs, emotions, \u0026amp; environmental interaction of virtual world from the real world, thus transforming the experience of user in the intelligent healthcare, human-machine interaction, AR, and VR [virtual reality].\u003c/p\u003e \u003cp\u003eThe technology of AR offers an auspicious choice for enriching the sensorial views \u0026amp; the interaction needs [Morris \u0026amp; Yeboah, 2023]. The AR system supplements the reality, that could not be embodied sufficiently in a label of object by means of superimposing virtual objects [computer-generated], like texts, graphics, and sounds over the real-world environment of the user. This could be imagined how the positioning of smart devices are facilitated by IoT devices \u0026amp; contents of AR [Joko, 2023] which might be aided from the visualization and interpretation of data with which the physician might interact with targeted health status and studies daily records over virtual contents. With the intention of using AR technology integrated with IoT/IoMT device for supporting visualization of data, in this manuscript, a novel design framework which integrated IoT with AR dependent framework termed AR-IoT is proposed. The main intention of this model is to design a framework to track employee\u0026rsquo;s healthcare status such as heart beat and brain health with the use of IoT devices. The health status monitoring is integrated with Augmented reality by employing DL models at which the recognition or tracking of normal and abnormal health status of employees are carried by processing the input data from sensors using Artificial Intelligence and Deep Learning models. From this, an alert message will be sent to the physician while abnormal tracking occurs which could be monitored and given necessary actions via Augmented reality technology.\u003c/p\u003e \u003cp\u003eThe residual sections of this manuscript are segregated as follows: section II is the related works offered related to this AR based IoT technology. The proposed design framework is narrated in section III. The analysis of performance made for proposed model is shown in section IV. The conclusions are provided in section V.\u003c/p\u003e"},{"header":"2 Related Work","content":"\u003cp\u003eA short review of various conventional models related to AR technology integrated with IoT is projected in this section.\u003c/p\u003e \u003cp\u003eThe author of the work [Aditya \u0026amp; Dash, 2022] aims to implement IoT integration with the AR model to visualize and assess virtual data over real-time objects in a convenient manner. The step-wise process was given like how one could employ AR and IoT altogether which was not much narrated in review. The AR was employed for viewing data over real-world things. The mobile camera captures \u0026amp; records the video thus employing Vuforia SDK of any virtual data such as 3D shapes, and environments which could be overlaid in video. Therefore, the virtual content that comprises of 3D model of the machine together with sensor data could be shown in the real-world video.\u003c/p\u003e \u003cp\u003eIt was stated in [Gomes et al., 2020], that AR was the digital content overlay in the real-time world which thus plays a significant role in the product's lifecycle from their support to design, thereby permitting high flexibility. The interconnected system adoption and the utilization of IoT have driven the utilization of AI technology due to excess data that comes from varied sources will be unstructured. Various AI-based models were employed for decades thus aiming to make sense of unstructured data thus transforming them to suitable data. Hence, the AR, converging IoT, and AI thus makes the system autonomous enchantingly and a problem-solving one in various conditions.\u003c/p\u003e \u003cp\u003eA study [Ghorbani, 2024] reported the outcome of the assistive intelligence system [IA], thus attained over the IoT integration, AR, \u0026amp; adaptive fuzzy decision-making models. The suggested model has 4 major mechanisms the location and stored heading data at the local fog layer, the AR device that interacts with the patients, a supervisory decision-making module for handling the environmental and direct interactions with patients, and the user interface for caregivers or family for monitoring the real-time situation of patient and sends reminders or alerts if needed.\u003c/p\u003e \u003cp\u003eA new architecture on IAQ which was based on cognitive IoT [CIoT] with the use of AR was presented in the work [Sassi \u0026amp; Fourati, 2020]. The suggested model offers huge potential for integrating the IoT for the data of IAQ with the visualization of AR. This in turn allows instinctive communication between users \u0026amp; the devices of IoT which enhances the data visualization further that were composed of sensors of IoT that contribute effectively to improve the usability and experience of user.\u003c/p\u003e \u003cp\u003eIn the work [Kim et al., 2021], a comprehensive review was made of various proposed MI models which use AR and IoT. About 23 studies in total were analyzed and identified over the rigorous review protocol. The outcome of this MI system architecture model, their relationship among system components, and interaction modalities of input/output. Their open research challenges were discussed and presented for summarizing the results and thus identifies future development of research paths for MI developers and researchers.\u003c/p\u003e \u003cp\u003eA system was proposed in the work [Naheem et al., 2023] at which the crucial data for physicians was superimposed at its real-time surroundings through the semi-transparent glasses housed in the AR headset. In this, sensors that were worn by the patients in hospitals collect information of real-worlds, thus processing them and sending them to the doctors' AR glasses through wireless for altering them regarding any such abnormality state of the patient. Depending on the present health status of the patient, the doctor could take necessary actions to improve the condition of the patient.\u003c/p\u003e \u003cp\u003eA prototype platform for exergames was presented in the work [Koulouris et al., 2022] which integrated AR with IoT on the commodity of mobile devices to develop serious games regarding healthcare field. The major intention of this solution was to augment the use of gamification models for boosting the physical activities of the patient and for assisting them regarding the usual assessment of health and cognitive status over challenges \u0026amp; quests in a real and virtual world. A solution was thus validated in real-time scenarios and outcomes were interpreted to enhance the usability and performance of the prototype.\u003c/p\u003e \u003cp\u003eThe work [Elavarasi \u0026amp; Kavitha, 2023] presents, progresses, and thus authenticates AR \u0026amp; IoT-aided systems of healthcare to be employed by physicians. The suggested model was based on a smart city IoT-middleware structure offering an intuitive, standardized, and non-intrusive manner of delivering data on elder persons to their clinicians. This prototype was presented thus assessing the outcome which reveals the system's efficiency. The time of execution on average includes the detection of objects, thus communicating them with the server, and offering outcomes in the application.\u003c/p\u003e \u003cp\u003eThe suggested framework of AR/MR-IoT makes use of open-source and standard protocols like HTTPS, MQTT, or Node-RED [Blanco-Novoa et al., 2020]. The review made in [Sahu et al., 2024] sheds light on various components needed for improving the virtual and augmented reality-based technology which was followed by various devices employed in virtual and augmented-based systems. The usage of this model thereby signifies the neural flaws \u0026amp; presented approaches for employing those advancements thus offering an immersive experience for both educational and research purposes. Hence, the role of IoT with VR and AR soon is to diagnose and treat neurological disorders.\u003c/p\u003e \u003cp\u003eAR is a promising technology as stated in [Aanjanadevi et al., 2022] which was the emerging key for unlocking the potential of IoT to the fullest. The challenges and potentials of integrating the IoT with AR were investigated here. After thorough process of searching, about 51 publications were selected and examined thoroughly for summarizing the recent developments more recently at this area. In the research identification, about 4 major clusters of potential advantages were found: sensor data visualization, interaction of human-object, diversity of application, and use case adoption of IoT. On the contrary, the limitations posed by AR and IoT integrations were thus clustered over organizational, technical, and ergonomic deliberations.\u003c/p\u003e \u003cp\u003eThe aim of this work [Siang et al., 2023] was to develop smart manufacturing plant integrated with AR and IoT interface so as to augment the controlling and monitoring process efficiency. At this work, an ASRS-based system was implemented and thus integrated with IoT and AR applications. The concluding prototype of this AR application allows user to monitor, control ASRS system in real-world. By the implementation of IoT and AR, efficiency of new ASRS system might augment hugely which enhances the control, monitor, and technical guidance support of system. Also, the interface of IoT might emerge for storing data or information related to the system, for instance, system status and storage information.\u003c/p\u003e"},{"header":"3 Proposed Work","content":"\u003cp\u003eThe proposed framework involves two stages the transmitter and receiver side. At the side of the transmitter, the sensors are connected with the Raspberry Pi module which is connected to the camera and Wi-Fi module. The output from this is transmitted to the receiver side where the input data are recognized and classified for any unstable healthcare condition of the employees or patients which could be connected and monitored over the Augmented reality.\u003c/p\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e3.1 Hardware sensors and data acquisition from the transmitter\u003c/h2\u003e \u003cp\u003ePresently, wearable sensors are regarded as a new tool for human activity recognition [HAR], placed in various parts of the human body. This is significant for knowing the precise body location of the wearable sensors and is the precise tool for attaching them to the human body. The location of the sensor on the body has an important impact on estimating the movements of the body and identifying the activities of the body, such that research is done in this field. The sensors could be placed and are visible and located typically on the sternum, waist, and belt. The wearable sensors over the placement of waist could monitor the movements of humans more precisely as this is nearer to the center of human body. The sensor numbers like the location of the sensor in HARS are needed. As per the research, the integration of thighs, chest, and ankles are embedded in the sensors and is more precise. Most often, smartphones incorporate entire kinds of sensors. The body temperature sensor, blood pressure, Heart rate sensor, Sp02 sensor, Blood Glucose sensor, and activity monitoring are placed in wearable device. The people employ wearable sensors for generating excess information regarding its movement, position, interaction, and location. The peripheral sensors might attain information regarding the environment of smart home like humidity, temperature, pressure, sound, light, and so on. They are not created for monitoring the activities in groups and thus discriminating them among residents\u0026rsquo; actions or movements. Various sensors could perform several tasks on monitoring for measuring properties like position, movement, ECG, and temperature. The camera is likewise a traditional tool for carefully attaining information. Adequate 2-D data is thus provided from various viewpoints to extract 3-D movements on humans and thus the environments are pre-determined. This view of the area on fixed cameras is thus\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003elimited. Other limitation of cameras covers the fact that several people does not feel gratified that entire movements are under control. The wireless communication is made using a Wi-Fi module for transmitting the health care data to the IoT cloud module. Figure\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ea. Defines the sensors are connected to the camera and Wi-Fi module using a Raspberry Pi module and b. defines the AR with receiver output.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003e3.2 Input Pre-processing\u003c/h2\u003e \u003cp\u003eThe input healthcare data of employees exhibits the subsequent artifacts: this is not complete and thus lacks fewer essential attributes or this must include collective data, as it is noisy and has some errors, this is not a consistent one and comprises similar codes or naming discrepancies. For carrying out such an analysis this is essential that a data must undergoes the pre-processing stage which covers cleaning, integration, transformation, reduction, and discretization. For carrying cleaning, the attributes that are missing in input data is thus filled thus identifying outliers. The noisy data is thus smoothened to rectify inconsistent data. The process of data transformation covers data normalization, aggregation, generalization, and attribute generation. In data reduction, attributes number and tuples are thus decreased together with the data number of attributes. So as to attain this, contours that were probable with specifications were identified for detecting abnormalities.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003e3.3 EM Based Clustering\u003c/h2\u003e \u003cp\u003eEM is the probabilistic and iterative approach that switches among the expectation [E] and maximization [m] stages consecutively. In E-phase, EM computes the likelihood function of expected values. In the M-phase, moreover, EM attains parameter estimation for maximizing the likelihood function. The attained parameters at the M-phase were employed at the subsequent E-phase. This process gets repeated till there is an occurrence of convergence that is it convergences to the parameter\u0026rsquo;s final values. For performing E-phase at the estimation of probabilities of the input data which belongs to cluster or model mixture, EM carried subsequent formula:\u003cdiv id=\"Equ1\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ1\" name=\"EquationSource\"\u003e\n$$\\:{\\varvec{p}\\left(\\varvec{m}\\vee\\:\\varvec{n}\\right)}^{\\varvec{t}}=\\frac{{\\varvec{w}}_{\\varvec{c}}^{\\varvec{t}}\\varvec{h}\\left({\\varvec{X}}_{\\varvec{n}},{\\varvec{\\mu\\:}}_{\\varvec{m}}^{\\varvec{t}},{\\varvec{\\sigma\\:}}_{\\varvec{m}}^{\\varvec{t}}\\right)}{{\\sum\\:}_{\\varvec{k}=1}^{\\varvec{M}}{\\varvec{w}}_{\\varvec{c}}^{\\varvec{t}}\\varvec{h}\\left({\\varvec{X}}_{\\varvec{n}},{\\varvec{\\mu\\:}}_{\\varvec{m}}^{\\varvec{t}},{\\varvec{\\sigma\\:}}_{\\varvec{m}}^{\\varvec{t}}\\right)}$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e1\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003eHere, h signifies the function of the probability density of the input \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{X}_{n}\\)\u003c/span\u003e\u003c/span\u003e at dataset for m cluster having standard deviation \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\sigma\\:}_{m}^{t}\\)\u003c/span\u003e\u003c/span\u003e, iteration t, and \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\mu\\:}_{m}^{t}\\)\u003c/span\u003e\u003c/span\u003e mean. In this, the normal univariant, multivariate, or bivariate function of probability density might be utilized and this is according to the data dimensionality. Also, at the constraint \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\sum\\:}_{k=1}^{M}{w}_{k}^{t}=1\\)\u003c/span\u003e\u003c/span\u003e, data allocation for the mixture models is then influenced through the weighting factor \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{w}^{t}\\)\u003c/span\u003e\u003c/span\u003e. As per the M-phase performance, for the likelihood maximization function, EM should need the computation of subsequent each cluster estimation:\u003cdiv id=\"Equ2\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ2\" name=\"EquationSource\"\u003e\n$$\\:{\\mu\\:}_{m}^{t+1}=\\frac{{\\sum\\:}_{n=1}^{N}p{\\left(m\\vee\\:n\\right)}^{t}{x}_{n}}{{\\sum\\:}_{n=1}^{N}p{\\left(m\\vee\\:n\\right)}^{t}}$$\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$$\\:{\\sigma\\:}_{m}^{t+1}=\\sqrt{\\frac{{\\sum\\:}_{n=1}^{N}p{\\left(m\\vee\\:n\\right)}^{t}{x}_{n}-{\\mu\\:}_{m}^{t+1}}{{\\sum\\:}_{n=1}^{N}p{\\left(m\\vee\\:n\\right)}^{t}}}$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e3\u003c/div\u003e\u003c/div\u003e\u003cdiv id=\"Equ4\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ4\" name=\"EquationSource\"\u003e\n$$\\:{w}_{k}^{t+1}=\\frac{{\\sum\\:}_{n=1}^{N}p{\\left(m\\vee\\:n\\right)}^{t}}{N}$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e4\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003eThe above process gets repeated on executing M and E phases till the occurrence of convergence.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003e3.4 Feature extraction and optimal selection of features\u003c/h2\u003e \u003cp\u003eAfter the clustering process, the extraction of features is carried out to mine relevant and suitable features to help the classification process. It is necessary to employ an optimal feature selection module to select best optimal feature subset for which Intelligent swarm dependent BAT optimization process is carried. This proposed Intelligent swarm dependent BAT algorithm is a method of optimization working based on intelligence of bats. There is the emission of short pulses by bat for their hunting and movement process. The pattern of successive navigation depends on echo returned. The bats could predict the obstacle kind depending on the signal of echo.\u003c/p\u003e \u003cp\u003e\u0026bull; The entire bats thus utilize the escalation process for predicting the distance \u0026amp; could differentiate between the obstacles and prey.\u003c/p\u003e \u003cp\u003e\u0026bull; Bats could fly with the speed of v\u003csub\u003ei\u003c/sub\u003e having frequency of f\u003csub\u003emin,\u003c/sub\u003e with position \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{x}_{i}\\)\u003c/span\u003e\u003c/span\u003e, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{A}_{o}\\)\u003c/span\u003e\u003c/span\u003e loudness \u0026amp; varying the wavelength λ to identify prey.\u003c/p\u003e \u003cp\u003e\u0026bull; Bats could be able to update the frequency of their expelled pulse and emission of pulse rate which is updated \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:r\\)\u003c/span\u003e\u003c/span\u003e \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\in\\:\\left[\\text{0,1}\\right]\\)\u003c/span\u003e\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e\u0026bull; Consider that the loudness varies from the maximum \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{A}_{o}\\)\u003c/span\u003e\u003c/span\u003e value to the smaller \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{A}_{min}\\)\u003c/span\u003e\u003c/span\u003evalue\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:.\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003cp\u003eThe velocities and positions of the bat were initialized randomly. The frequency of pulse emission is altered \u0026amp; the equation is shown by:\u003cdiv id=\"Equa\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equa\" name=\"EquationSource\"\u003e\n$$\\:{f}_{i}={f}_{min}+\\left({f}_{max}-{f}_{min}\\right)\\beta\\:$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003eThe velocity is thus altered \u0026amp; their equation is provided by:\u003cdiv id=\"Equb\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equb\" name=\"EquationSource\"\u003e\n$$\\:{v}_{i}^{t}={v}_{i}^{t-1}+\\left({x}_{i}^{t}-{x}_{gbest}\\right){f}_{i}$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003eThe bat position is thus altered and their equation is given by:\u003cdiv id=\"Equc\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equc\" name=\"EquationSource\"\u003e\n$$\\:{x}_{i}^{t}={x}_{i}^{t-1}+{v}_{i}^{t}$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003e \u003cspan class=\"InlineEquation\"\u003e \u003cspan class=\"mathinline\"\u003e\\(\\:{f}_{min}\\)\u003c/span\u003e \u003c/span\u003e\u0026ndash;\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:minimumfrequency\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003cp\u003e \u003cspan class=\"InlineEquation\"\u003e \u003cspan class=\"mathinline\"\u003e\\(\\:{f}_{max}\\)\u003c/span\u003e \u003c/span\u003e \u0026ndash; \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:maximumfrequency\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003cp\u003e \u003cspan class=\"InlineEquation\"\u003e \u003cspan class=\"mathinline\"\u003e\\(\\:{v}_{i}^{t-1}\\)\u003c/span\u003e \u003c/span\u003edenotes the preceding velocity \u0026amp; \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{x}_{gbest}\\)\u003c/span\u003e\u003c/span\u003e denotes the optimal position.\u003c/p\u003e \u003cp\u003eA newer solution is updated for its local position and is provided by the equation:\u003cdiv class=\"BlockQuote\"\u003e\u003cp\u003e \u003cspan class=\"InlineEquation\"\u003e \u003cspan class=\"mathinline\"\u003e\\(\\:{x}_{update}\\)\u003c/span\u003e \u003c/span\u003e = \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{x}_{old}\\)\u003c/span\u003e\u003c/span\u003e + \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\in\\:{A}^{t}\\)\u003c/span\u003e\u003c/span\u003e (8)\u003c/p\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003e \u003cspan class=\"InlineEquation\"\u003e \u003cspan class=\"mathinline\"\u003e\\(\\:\\in\\:\\:\\)\u003c/span\u003e \u003c/span\u003esignifies the random number among [-1,1].\u003c/p\u003e \u003cp\u003eThe algorithm for this intelligent swarm-dependent BAT algorithm is provided below:\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"No\" id=\"Taba\" border=\"1\"\u003e \u003ccolgroup cols=\"1\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAlgorithm 1: Intelligent swarm dependent BAT optimization approach\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 class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\text{I}\\text{n}\\text{i}\\text{t}\\text{i}\\text{a}\\text{l}\\text{i}\\text{z}\\text{e}\\:\\text{b}\\text{a}\\text{t}\\text{p}\\text{o}\\text{p}\\text{u}\\text{l}\\text{a}\\text{t}\\text{i}\\text{o}\\text{n}\\)\u003c/span\u003e\u003c/span\u003e as \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\text{x}}_{\\text{i}}\\)\u003c/span\u003e\u003c/span\u003e, i\u0026thinsp;=\u0026thinsp;1,2,\u0026hellip;\u0026hellip;..n \u0026amp; frequencies\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\text{f}}_{\\text{i}},\\text{s}\\text{p}\\text{e}\\text{e}\\text{d}{\\text{v}}_{\\text{i}}\\)\u003c/span\u003e\u003c/span\u003e, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\text{l}\\text{o}\\text{u}\\text{d}\\text{n}\\text{e}\\text{s}\\text{s}\\)\u003c/span\u003e\u003c/span\u003e \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\text{A}}_{\\text{i}}\\)\u003c/span\u003e\u003c/span\u003e \u0026amp; rate of pulse \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\text{r}}_{\\text{i}}\\)\u003c/span\u003e\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\text{w}\\text{h}\\text{i}\\text{l}\\text{e}\\:\\left(\\text{t}\u0026lt;\\text{m}\\text{a}\\text{x}\\text{i}\\text{m}\\text{u}\\text{m}\\:\\text{i}\\text{t}\\text{e}\\text{r}\\text{a}\\text{t}\\text{i}\\text{o}\\text{n}\\text{s}\\right)\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\text{P}\\text{r}\\text{o}\\text{d}\\text{u}\\text{c}\\text{e}\\:\\text{n}\\text{e}\\text{w}\\:\\text{s}\\text{o}\\text{l}\\text{u}\\text{t}\\text{i}\\text{o}\\text{n}\\text{s}\\:\\text{b}\\text{y}\\:\\text{a}\\text{l}\\text{t}\\text{e}\\text{r}\\text{i}\\text{n}\\text{g}\\:\\text{t}\\text{h}\\text{e}\\:\\text{f}\\text{r}\\text{e}\\text{q}\\text{u}\\text{e}\\text{n}\\text{c}\\text{y}\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\text{M}\\text{o}\\text{d}\\text{i}\\text{f}\\text{y}\\:\\text{t}\\text{h}\\text{e}\\:\\text{p}\\text{o}\\text{s}\\text{i}\\text{t}\\text{i}\\text{o}\\text{n}\\text{s}\\wedge\\:\\text{v}\\text{e}\\text{l}\\text{o}\\text{c}\\text{i}\\text{t}\\text{i}\\text{e}\\text{s}\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\text{i}\\text{f}\\)\u003c/span\u003e\u003c/span\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\text{r}}_{\\text{i}}\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\text{C}\\text{h}\\text{o}\\text{o}\\text{s}\\text{e}\\:\\text{t}\\text{h}\\text{e}\\:\\text{o}\\text{p}\\text{t}\\text{i}\\text{m}\\text{a}\\text{l}\\:\\text{s}\\text{o}\\text{l}\\text{u}\\text{t}\\text{i}\\text{o}\\text{n}\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\text{C}\\text{r}\\text{e}\\text{a}\\text{t}\\text{e}\\:\\text{a}\\:\\text{l}\\text{o}\\text{c}\\text{a}\\text{l}\\:\\text{s}\\text{o}\\text{l}\\text{u}\\text{t}\\text{i}\\text{o}\\text{n}\\:\\text{a}\\text{r}\\text{o}\\text{u}\\text{n}\\text{d}\\:\\text{t}\\text{h}\\text{e}\\:\\text{o}\\text{p}\\text{t}\\text{i}\\text{m}\\text{a}\\text{l}\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\text{e}\\text{n}\\text{d}\\text{i}\\text{f}\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\text{C}\\text{r}\\text{e}\\text{a}\\text{t}\\text{e}\\:\\text{a}\\:\\text{o}\\text{r}\\text{i}\\text{g}\\text{i}\\text{n}\\text{a}\\text{l}\\:\\text{s}\\text{o}\\text{l}\\text{u}\\text{t}\\text{i}\\text{o}\\text{n}\\:\\text{b}\\text{y}\\:\\text{r}\\text{a}\\text{n}\\text{d}\\text{o}\\text{m}\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\text{i}\\text{f}\\)\u003c/span\u003e\u003c/span\u003e[random \u0026lt; \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\text{A}}_{\\text{i}}\\)\u003c/span\u003e\u003c/span\u003e] and [f[\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\text{x}}_{\\text{i}}\u0026lt;\\)\u003c/span\u003e\u003c/span\u003e f[\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\text{x}}_{\\text{g}\\text{b}\\text{e}\\text{s}\\text{t}}]\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\text{i}\\text{d}\\text{e}\\text{n}\\text{t}\\text{i}\\text{f}\\text{y}\\:\\text{t}\\text{h}\\text{e}\\:\\text{o}\\text{r}\\text{i}\\text{g}\\text{i}\\text{n}\\text{a}\\text{l}\\:\\text{s}\\text{o}\\text{l}\\text{u}\\text{t}\\text{i}\\text{o}\\text{n}\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\text{U}\\text{p}\\text{d}\\text{a}\\text{t}\\text{e}\\:\\text{b}\\text{y}\\:\\text{i}\\text{n}\\text{c}\\text{r}\\text{e}\\text{a}\\text{s}\\text{i}\\text{n}\\text{g}\\)\u003c/span\u003e\u003c/span\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\text{r}}_{\\text{i}}\\)\u003c/span\u003e\u003c/span\u003e\u0026amp; \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\text{d}\\text{e}\\text{c}\\text{r}\\text{e}\\text{a}\\text{s}\\text{e}\\)\u003c/span\u003e\u003c/span\u003e \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\text{A}}_{\\text{i}}\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\text{e}\\text{n}\\text{d}\\text{i}\\text{f}\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\text{F}\\text{i}\\text{n}\\text{d}\\:\\text{t}\\text{h}\\text{e}\\:\\text{c}\\text{u}\\text{r}\\text{r}\\text{e}\\text{n}\\text{t}\\:\\text{b}\\text{e}\\text{s}\\text{t}\\wedge\\:\\text{r}\\text{a}\\text{n}\\text{k}\\:\\text{t}\\text{h}\\text{e}\\:\\text{b}\\text{a}\\text{t}\\text{s}\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\text{e}\\text{n}\\text{d}\\text{w}\\text{h}\\text{i}\\text{l}\\text{e}\\)\u003c/span\u003e\u003c/span\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 \u003cp\u003eConsequently, the optimal range of features are extracted and selected utilizing this proposed optimization approach.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003e3.5 Hybrid Inception v3 and MobileNet v2 Classifier model\u003c/h2\u003e \u003cp\u003eThe presented hybrid MobileNet v2 and Inception v3 model is utilized to predict the classification process. Figure\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e. Defines the MobileNet v2 structure consists of the convolutional layers at each of them is then followed by the batch normalization layer and ReLU non-linear activation function which excludes the output layer. The convolutional layers are expensive in the MobileNet V2 which is factorized as the lightweight convolutional blocks which are separable. The input block is filtered at each block over the depth-wise convolutional layer having size [3x3]. The output channels offered are thus projected over the pointwise convolutional layer having size [1x1]. The MobileNet v2 blocks are built usually with residual connectivity that helps converge network weight. The depth wise utilization with 3x3 removeable convolutions \u0026amp; the resolutions modification over layers infers on decreasing the of computation complexity that corresponds to the standardized form of convolutions despite minor variation in accuracy.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eFor the image classification, Inception V3 is likewise employed which is extended typically from the network of GoogleNet expressed in Figure. 3. This relies on offering several smaller convolutions having similar level instead of employing huge convolutional ranges. The first layer of network consists of three convolutions and one max-pooling layer. The final one comprises of channels that are provided and are merged non-linearly. Therefore, the network parameter reduction is carried which implies on the training or testing accelerations stages. Likewise, the features are extracted effectively which increases the classification accuracy. This model consists of dense layers to increase the network depth that reduces complexity of computation once more. This technique provides higher rate of accuracy than other traditional techniques like ResNet, AlexNet, and GoogLeNet. Due to its flexibility and accuracy, this technique offers higher performance.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe suggested model employs score-dependent functions for estimating score values \u0026amp; creates prediction effectively. The computation of score value is made by means of following equation:\u003cdiv id=\"Equ5\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ5\" name=\"EquationSource\"\u003e\n$$\\:score={\\sum\\:}_{i=1}^{M}{\\sum\\:}_{j=1}^{N}\\left({A}_{1}\\left(i,j\\right)+{A}_{2}\\left(i,j\\right)\\right)/2$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e9\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003eIn this, A denotes the score generated by classifier model presented. N signifies number of classes \u0026amp; M denotes test sets. On employing above equation, new score having M\u0026times;N size is achieved. After that, using this new score value, class labels are determined as per the maximum value of each line [M]. The algorithm for this is shown below:\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"No\" id=\"Tabb\" border=\"1\"\u003e \u003ccolgroup cols=\"1\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAlgorithm 2: Proposed Hybrid classification technique\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 class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\text{I}\\text{n}\\text{p}\\text{u}\\text{t}:\\text{s}\\text{c}\\text{o}\\text{r}\\text{e}1,\\text{s}\\text{c}\\text{o}\\text{r}\\text{e}2,\\text{Y}\\text{T}\\text{e}\\text{s}\\text{t}\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\text{O}\\text{u}\\text{t}\\text{p}\\text{u}\\text{t}:\\text{a}\\text{c}\\text{c}\\text{u}\\text{r}\\text{a}\\text{c}\\text{y}\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\text{S}\\text{t}\\text{e}\\text{p}1:\\:\\text{f}\\text{o}\\text{r}\\:\\text{i}=1\\:\\text{t}\\text{o}\\:\\text{M}\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\text{S}\\text{t}\\text{e}\\text{p}2:\\)\u003c/span\u003e\u003c/span\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\text{f}\\text{o}\\text{r}\\:\\text{j}=1\\:\\text{t}\\text{o}\\:\\text{N}\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\text{S}\\text{t}\\text{e}\\text{p}3\\)\u003c/span\u003e\u003c/span\u003e: \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\text{n}\\text{e}\\text{w}\\text{e}\\text{r}}_{\\text{s}\\text{c}\\text{o}\\text{r}\\text{e}}\\left(\\text{i},\\text{j}\\right)=\\left(\\text{s}\\text{c}\\text{o}\\text{r}\\text{e}1\\left(\\text{i},\\text{j}\\right)+\\text{s}\\text{c}\\text{o}\\text{r}\\text{e}2\\left(\\text{i},\\text{j}\\right)\\right)/2\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\text{S}\\text{t}\\text{e}\\text{p}4:\\)\u003c/span\u003e\u003c/span\u003e \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\text{e}\\text{n}\\text{d}\\text{f}\\text{o}\\text{r}\\:\\text{j}\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\text{S}\\text{t}\\text{e}\\text{p}5:\\)\u003c/span\u003e\u003c/span\u003e \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\text{e}\\text{n}\\text{d}\\text{f}\\text{o}\\text{r}\\:\\text{i}\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\text{S}\\text{t}\\text{e}\\text{p}6:\\:\\text{f}\\text{o}\\text{r}\\:\\text{k}=1\\text{t}\\text{o}\\:\\text{K}\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\text{S}\\text{t}\\text{e}\\text{p}7:\\)\u003c/span\u003e\u003c/span\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\text{i}\\text{f}\\:\\text{s}\\text{c}\\text{o}\\text{r}\\text{e}\\left(\\text{k},1\\right)\\ge\\:\\text{s}\\text{c}\\text{o}\\text{r}\\text{e}\\left(\\text{i},2\\right)\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\text{S}\\text{t}\\text{e}\\text{p}8:\\)\u003c/span\u003e\u003c/span\u003e \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\text{Y}\\text{P}\\text{r}\\text{e}\\text{d}\\left(\\text{k}\\right)=1;\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\text{S}\\text{t}\\text{e}\\text{p}9:\\)\u003c/span\u003e\u003c/span\u003e \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\text{e}\\text{l}\\text{s}\\text{e}\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\text{S}\\text{t}\\text{e}\\text{p}11:\\)\u003c/span\u003e\u003c/span\u003e \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\text{Y}\\text{P}\\text{r}\\text{e}\\text{d}\\left(\\text{k}\\right)=1;\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\text{S}\\text{t}\\text{e}\\text{p}12:\\:\\text{a}\\text{c}\\text{c}\\text{u}\\text{r}\\text{a}\\text{c}\\text{y}=\\text{m}\\text{e}\\text{a}\\text{n}\\left(\\text{Y}\\text{p}\\text{r}\\text{e}\\text{d}=\\text{Y}\\text{T}\\text{e}\\text{s}\\text{t}\\right);\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003cp\u003eS\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\text{t}\\text{e}\\text{p}13:\\)\u003c/span\u003e\u003c/span\u003e \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\text{e}\\text{n}\\text{d}\\text{f}\\text{o}\\text{r}\\:\\text{k}\\)\u003c/span\u003e\u003c/span\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 \u003cp\u003eThe Hybrid proposed classifier model predicts the employee\u0026rsquo;s healthcare abnormality and offers classification outcome with high range of accuracy. This performance is estimated and is provided in subsequent section.\u003c/p\u003e \u003c/div\u003e"},{"header":"4 The Performance Analysis","content":"\u003cp\u003eThe perpetuation of experiments is evaluated for the suggested model with varied instance numbers that range from 10% to 100% and are compared with traditional models like LSTM, FIS integrated with LSTM [FLSTM], and Bi-LSTM [Nancy et al., 2022]. The comparison is made to show the effectiveness of the proposed model over other existing models related to the most widely used approach LSTM.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eTable 1: Comparative analysis of accuracy\u003c/em\u003e\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"294\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eData [%]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eBi-LSTM\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e94.04\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e94.56\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e94.86\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e96.87\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e20\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e94.35\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e95.59\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e96.02\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e97.98\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e30\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e94.52\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e96.21\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e96.76\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e98.54\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e40\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e94.68\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e96.68\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e97.25\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e98.90\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e50\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e94.75\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e97.01\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e97.65\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e98.95\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e60\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e94.86\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e97.31\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e97.96\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e98.99\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e70\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e94.89\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e97.51\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e98.24\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e99.04\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e80\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e94.95\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e97.70\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e98.48\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e99.65\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e90\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e95.00\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e97.92\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e98.67\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e99.73\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e100\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e95.07\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e98.04\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e98.86\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e99.89\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eTable 1 shows the comparative analysis of accuracy for varied number of data for both proposed and existing models. The analysis estimated reveals that the proposed model is better and is enhanced than other traditional models. The graphical representation of this is shown in figure 4.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eTable 2 signifies the comparative examination of precision for the varied number of data for both proposed and existing models. The analysis estimated reveals that the suggested technique\u0026rsquo;s precision is augmented and is better than other traditional models. The graphical representation of this is shown in figure 5.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eTable 2: Comparative analysis of precision\u003c/em\u003e\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 61px;\"\u003e\n \u003cp\u003eData [%]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 61px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 61px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 61px;\"\u003e\n \u003cp\u003eBi-LSTM\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 61px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e94.00\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e94.60\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e94.81\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e96.42\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e20\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e94.37\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e95.55\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e96.00\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e97.28\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e30\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e94.52\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e96.21\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e96.74\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e98.33\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e40\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e94.67\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e96.69\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e97.22\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e98.49\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e50\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e94.72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e97.04\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e97.69\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e98.50\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e60\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e94.86\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e97.30\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e97.95\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e98.86\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e70\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e94.85\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e97.50\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e98.21\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e99.01\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e80\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e94.99\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e97.71\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e98.50\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e99.35\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e90\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e95.00\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e97.95\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e98.63\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e99.73\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e100\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e95.07\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e98.03\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e98.90\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e99.88\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eTable 3 represents the comparative analysis of recall for varied number of data for both proposed and existing models. The analysis estimated reveals that the proposed model recall is better and is enhanced than other traditional models. The graphical representation of this is shown in figure 6.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eTable 3: Comparative analysis of recall\u003c/em\u003e\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"295\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 43px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eData [%]\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 54px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eLSTM\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 63px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eFLSTM\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 65px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eBi-LSTM\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 71px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eProposed\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 43px;\"\u003e\n \u003cp\u003e10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 54px;\"\u003e\n \u003cp\u003e94.08\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 63px;\"\u003e\n \u003cp\u003e94.51\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 65px;\"\u003e\n \u003cp\u003e94.90\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 71px;\"\u003e\n \u003cp\u003e95.11\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 43px;\"\u003e\n \u003cp\u003e20\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 54px;\"\u003e\n \u003cp\u003e94.32\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 63px;\"\u003e\n \u003cp\u003e95.63\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 65px;\"\u003e\n \u003cp\u003e96.04\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 71px;\"\u003e\n \u003cp\u003e96.27\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 43px;\"\u003e\n \u003cp\u003e30\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 54px;\"\u003e\n \u003cp\u003e94.52\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 63px;\"\u003e\n \u003cp\u003e96.21\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 65px;\"\u003e\n \u003cp\u003e96.78\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 71px;\"\u003e\n \u003cp\u003e96.38\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 43px;\"\u003e\n \u003cp\u003e40\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 54px;\"\u003e\n \u003cp\u003e94.69\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 63px;\"\u003e\n \u003cp\u003e96.68\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 65px;\"\u003e\n \u003cp\u003e97.29\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 71px;\"\u003e\n \u003cp\u003e97.40\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 43px;\"\u003e\n \u003cp\u003e50\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 54px;\"\u003e\n \u003cp\u003e94.79\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 63px;\"\u003e\n \u003cp\u003e97.00\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 65px;\"\u003e\n \u003cp\u003e97.61\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 71px;\"\u003e\n \u003cp\u003e97.88\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 43px;\"\u003e\n \u003cp\u003e60\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 54px;\"\u003e\n \u003cp\u003e94.87\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 63px;\"\u003e\n \u003cp\u003e97.32\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 65px;\"\u003e\n \u003cp\u003e97.97\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 71px;\"\u003e\n \u003cp\u003e97.99\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 43px;\"\u003e\n \u003cp\u003e70\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 54px;\"\u003e\n \u003cp\u003e94.92\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 63px;\"\u003e\n \u003cp\u003e97.53\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 65px;\"\u003e\n \u003cp\u003e98.27\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 71px;\"\u003e\n \u003cp\u003e98.67\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 43px;\"\u003e\n \u003cp\u003e80\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 54px;\"\u003e\n \u003cp\u003e94.91\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 63px;\"\u003e\n \u003cp\u003e97.68\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 65px;\"\u003e\n \u003cp\u003e98.45\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 71px;\"\u003e\n \u003cp\u003e98.70\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 43px;\"\u003e\n \u003cp\u003e90\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 54px;\"\u003e\n \u003cp\u003e95.00\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 63px;\"\u003e\n \u003cp\u003e97.89\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 65px;\"\u003e\n \u003cp\u003e98.72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 71px;\"\u003e\n \u003cp\u003e99.63\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 43px;\"\u003e\n \u003cp\u003e100\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 54px;\"\u003e\n \u003cp\u003e95.06\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 63px;\"\u003e\n \u003cp\u003e98.03\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 65px;\"\u003e\n \u003cp\u003e98.81\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 71px;\"\u003e\n \u003cp\u003e99.91\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eTable 4 signifies the comparative analysis of specificity for varied data numbers for both proposed and existing models. The estimated analysis exposes that the proposed model specificity is better and is enhanced than other traditional models. The graphical depiction of this is revealed in Figure 7.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eTable 4: Comparative analysis of specificity\u003c/em\u003e\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"312\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 64px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eData [%]\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 56px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eLSTM\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eFLSTM\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 54px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eBi-LSTM\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 72px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eProposed\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 64px;\"\u003e\n \u003cp\u003e10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e94.00\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003e94.60\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 54px;\"\u003e\n \u003cp\u003e94.81\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 72px;\"\u003e\n \u003cp\u003e95.77\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 64px;\"\u003e\n \u003cp\u003e20\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e94.37\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003e95.56\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 54px;\"\u003e\n \u003cp\u003e96.01\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 72px;\"\u003e\n \u003cp\u003e96.43\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 64px;\"\u003e\n \u003cp\u003e30\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e94.52\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003e96.21\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 54px;\"\u003e\n \u003cp\u003e96.74\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 72px;\"\u003e\n \u003cp\u003e96.89\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 64px;\"\u003e\n \u003cp\u003e40\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e94.67\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003e96.69\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 54px;\"\u003e\n \u003cp\u003e97.22\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 72px;\"\u003e\n \u003cp\u003e97.51\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 64px;\"\u003e\n \u003cp\u003e50\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e94.72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003e97.04\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 54px;\"\u003e\n \u003cp\u003e97.68\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 72px;\"\u003e\n \u003cp\u003e97.74\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 64px;\"\u003e\n \u003cp\u003e60\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e94.86\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003e97.30\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 54px;\"\u003e\n \u003cp\u003e97.95\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 72px;\"\u003e\n \u003cp\u003e97.99\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 64px;\"\u003e\n \u003cp\u003e70\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e94.85\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003e97.50\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 54px;\"\u003e\n \u003cp\u003e98.21\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 72px;\"\u003e\n \u003cp\u003e98.61\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 64px;\"\u003e\n \u003cp\u003e80\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e94.99\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003e97.71\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 54px;\"\u003e\n \u003cp\u003e98.50\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 72px;\"\u003e\n \u003cp\u003e98.94\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 64px;\"\u003e\n \u003cp\u003e90\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e95.00\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003e97.95\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 54px;\"\u003e\n \u003cp\u003e98.63\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 72px;\"\u003e\n \u003cp\u003e99.36\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 64px;\"\u003e\n \u003cp\u003e100\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e95.07\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003e98.03\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 54px;\"\u003e\n \u003cp\u003e98.90\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 72px;\"\u003e\n \u003cp\u003e99.88\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eTable 5 indicates the comparative study of the F1-score for varied data numbers for both proposed and existing models. The analysis estimated exposes that the suggested scheme F1-score is improved and enhanced than other traditional models. The graphical illustration of this is provided in figure 8.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eTable 5. comparative analysis of F1-score\u003c/em\u003e\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"293\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 43px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eData [%]\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 54px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eLSTM\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 62px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eFLSTM\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 64px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eBi-LSTM\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 70px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eProposed\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 43px;\"\u003e\n \u003cp\u003e10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 54px;\"\u003e\n \u003cp\u003e94.04\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 62px;\"\u003e\n \u003cp\u003e94.56\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 64px;\"\u003e\n \u003cp\u003e94.85\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 70px;\"\u003e\n \u003cp\u003e95.88\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 43px;\"\u003e\n \u003cp\u003e20\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 54px;\"\u003e\n \u003cp\u003e94.35\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 62px;\"\u003e\n \u003cp\u003e95.59\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 64px;\"\u003e\n \u003cp\u003e96.02\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 70px;\"\u003e\n \u003cp\u003e96.47\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 43px;\"\u003e\n \u003cp\u003e30\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 54px;\"\u003e\n \u003cp\u003e94.52\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 62px;\"\u003e\n \u003cp\u003e96.21\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 64px;\"\u003e\n \u003cp\u003e96.76\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 70px;\"\u003e\n \u003cp\u003e96.90\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 43px;\"\u003e\n \u003cp\u003e40\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 54px;\"\u003e\n \u003cp\u003e94.68\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 62px;\"\u003e\n \u003cp\u003e96.68\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 64px;\"\u003e\n \u003cp\u003e97.25\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 70px;\"\u003e\n \u003cp\u003e97.52\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 43px;\"\u003e\n \u003cp\u003e50\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 54px;\"\u003e\n \u003cp\u003e94.75\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 62px;\"\u003e\n \u003cp\u003e97.02\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 64px;\"\u003e\n \u003cp\u003e97.65\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 70px;\"\u003e\n \u003cp\u003e97.84\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 43px;\"\u003e\n \u003cp\u003e60\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 54px;\"\u003e\n \u003cp\u003e94.86\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 62px;\"\u003e\n \u003cp\u003e97.31\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 64px;\"\u003e\n \u003cp\u003e97.96\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 70px;\"\u003e\n \u003cp\u003e98.09\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 43px;\"\u003e\n \u003cp\u003e70\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 54px;\"\u003e\n \u003cp\u003e94.88\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 62px;\"\u003e\n \u003cp\u003e97.51\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 64px;\"\u003e\n \u003cp\u003e98.24\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 70px;\"\u003e\n \u003cp\u003e98.75\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 43px;\"\u003e\n \u003cp\u003e80\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 54px;\"\u003e\n \u003cp\u003e94.95\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 62px;\"\u003e\n \u003cp\u003e97.70\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 64px;\"\u003e\n \u003cp\u003e98.48\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 70px;\"\u003e\n \u003cp\u003e98.79\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 43px;\"\u003e\n \u003cp\u003e90\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 54px;\"\u003e\n \u003cp\u003e95.00\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 62px;\"\u003e\n \u003cp\u003e97.92\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 64px;\"\u003e\n \u003cp\u003e98.67\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 70px;\"\u003e\n \u003cp\u003e99.03\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 43px;\"\u003e\n \u003cp\u003e100\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 54px;\"\u003e\n \u003cp\u003e95.07\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 62px;\"\u003e\n \u003cp\u003e98.03\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 64px;\"\u003e\n \u003cp\u003e98.86\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 70px;\"\u003e\n \u003cp\u003e99.15\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eTable 6 represents the overall comparative analysis of performance for both proposed and existing models in terms of \u0026nbsp; precision,. The estimated analysis shows that the proposed model outcome is better and is improved than other traditional models. The graphical illustration of this is provided in figure 9.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eTable 6: Overall comparison of performance\u003c/em\u003e\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"342\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 90px;\"\u003e\n \u003cp\u003e\u003cstrong\u003ePerformance Metrics\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 54px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eLSTM\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eFLSTM\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 60px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eBi-LSTM\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 72px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eProposed\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 90px;\"\u003e\n \u003cp\u003eAccuracy\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 54px;\"\u003e\n \u003cp\u003e95.07\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\n \u003cp\u003e98.04\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 60px;\"\u003e\n \u003cp\u003e98.86\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 72px;\"\u003e\n \u003cp\u003e99.89\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 90px;\"\u003e\n \u003cp\u003eSpecificity\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 54px;\"\u003e\n \u003cp\u003e95.07\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\n \u003cp\u003e98.03\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 60px;\"\u003e\n \u003cp\u003e98.90\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 72px;\"\u003e\n \u003cp\u003e99.88\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 90px;\"\u003e\n \u003cp\u003eRecall\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 54px;\"\u003e\n \u003cp\u003e95.06\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\n \u003cp\u003e98.04\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 60px;\"\u003e\n \u003cp\u003e98.81\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 72px;\"\u003e\n \u003cp\u003e99.91\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 90px;\"\u003e\n \u003cp\u003ePrecision\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 54px;\"\u003e\n \u003cp\u003e95.07\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\n \u003cp\u003e98.03\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 60px;\"\u003e\n \u003cp\u003e98.90\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 72px;\"\u003e\n \u003cp\u003e99.88\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 90px;\"\u003e\n \u003cp\u003eF1-score\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 54px;\"\u003e\n \u003cp\u003e95.07\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\n \u003cp\u003e98.03\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 60px;\"\u003e\n \u003cp\u003e98.86\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 72px;\"\u003e\n \u003cp\u003e99.15\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eTable 7 signifies the overall comparative performance analysis for both proposed and various existing models in terms of accuracy. The estimated analysis displays that the suggested scheme\u0026rsquo;s accuracy is higher \u0026amp; is augmented by comparing conventional models. https://www.kaggle.com/code/venkat1949/topic-modeling-based-biobert-qa-system. \u0026nbsp;\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eTable 7: Accuracy Comparison with existing models\u003c/em\u003e\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"374\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 276px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eModels\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 98px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAccuracy\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 276px;\"\u003e\n \u003cp\u003eEnsemble classifiers\u0026nbsp;[Latha \u0026amp; Jeeva, 2019]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003e85.4\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 276px;\"\u003e\n \u003cp\u003eHybrid RF and linear model\u0026nbsp;[Mohan et al., 2019]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003e88.70\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 276px;\"\u003e\n \u003cp\u003eType-2 fuzzy logic\u0026nbsp;[Long et al., 2015]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003e86.0\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 276px;\"\u003e\n \u003cp\u003eFuzzy analytic hierarchy \u0026amp; ANN\u0026nbsp;[Long et al., 2015]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003e91.0\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 276px;\"\u003e\n \u003cp\u003eStatistical model \u0026amp; DNN\u0026nbsp;[L. Ali et al., 2019]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003e91.57\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 276px;\"\u003e\n \u003cp\u003eAdaptive neuro-fuzzy\u0026nbsp;[Paul et al., 2018]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003e92.3\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 276px;\"\u003e\n \u003cp\u003eRelief feature selection \u0026amp; DT [28]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003e92.8\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 276px;\"\u003e\n \u003cp\u003e\u0026nbsp; [Ahmed et al., 2020]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003e94.15\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 276px;\"\u003e\n \u003cp\u003eCNN\u0026nbsp;[Kishore \u0026amp; Jayanthi, 2018]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003e97.0\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 276px;\"\u003e\n \u003cp\u003eFuzzy rules \u0026amp; DNN\u0026nbsp;[Mehmood et al., 2021]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003e96.5\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 276px;\"\u003e\n \u003cp\u003e\u0026nbsp; [Van Pham et al., 2018]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003e94.78\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 276px;\"\u003e\n \u003cp\u003eSequential forward selection \u0026amp; RF\u0026nbsp;[Dileep et al., 2023]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003e98.0\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 276px;\"\u003e\n \u003cp\u003eKernel RF\u0026nbsp;[Jabeen et al., 2019]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003e98.0\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 276px;\"\u003e\n \u003cp\u003eDCNN\u0026nbsp;[Muzammal et al., 2020]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003e98.2\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 276px;\"\u003e\n \u003cp\u003eEnsemble DL [36]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003e98.5\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 276px;\"\u003e\n \u003cp\u003eLinear SVC \u0026amp; DNN [37]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003e98.56\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 276px;\"\u003e\n \u003cp\u003eFuzzy information system \u0026amp; Bi-LSTM [21]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003e98.86\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 276px;\"\u003e\n \u003cp\u003eProposed\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003e99.89\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eFigure 10 signifies the graphical illustration of the overall comparative analysis of performance for both proposed and various existing schemes in terms of accuracy. The estimated analysis displays that the suggested model accuracy is better and is augmented over various existing methodologies.\u0026nbsp;\u003c/p\u003e"},{"header":"5 Conclusion","content":"\u003cp\u003eAn Augmented reality-based IoT model for monitoring or tracking employees\u0026rsquo; healthcare status was proposed in this work. The proposed design consists of transmitter and receiver side at which the transmitter side comprises of various sensors connected with Raspberry Pi module and is connected to camera and Wi-Fi module. The output from this is transmitted to the receiver side to process input cloud data for detecting abnormal health status using pre-processing, Expectation Maximization [EM] based clustering, and an Intelligent swarm-dependent BAT optimization approach to select the best features. and Hybrid Inception V3 and MobileNet V2 model for classification. In case, the abnormal status is detected, an alert will be sent to the caretaker or physician to take necessary actions. the performance estimation is made in terms of various performance metrics and the comparison made with the existing system reveals that the proposed model was improved over other existing models.\u003c/p\u003e"},{"header":"6 Future Work","content":"\u003cp\u003eThe challenges faced in such approaches are the security breaches like eavesdropping, jamming, denial of service [DOS] attacks, and spoofing attacks can be identified and control using proposed enhanced security with comprehensive future directions by creating IoB [Internet of Things Behaviour] to fight against attacks. The augmented approach gives a simulation approach to track the breaches that occurs while communicating with the devices.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003e“The authors declare that no funds, grants, or other support were received during the preparation of this manuscript.”\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting Interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003e“The authors have no relevant financial or non-financial interests to disclose.”\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor Contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003e“All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by Mr.G.Ganesan and Dr.S.Poonkuntran. The first draft of the manuscript was written by Mr.G.Ganesan and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.”\u003c/em\u003e\u003c/p\u003e\n\u003ch4\u003e\u003cstrong\u003eData Availability\u003c/strong\u003e\u003c/h4\u003e\n\u003cp\u003e\u003cem\u003e“The datasets generated during and/or analysed during the current study are not publicly available due to this work will be submitted for the author’s ph.d thesis but are available from the corresponding author on reasonable request..”\u003c/em\u003e\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003e[Aanjanadevi et al., 2022] Aanjanadevi, S., Palanisamy, V., Aanjankumar, S., Poonkuntran, S., \u0026amp; Karthikeyan, P. [2022]. Independent automobile intelligent motion controller and redirection, using a deep learning system. In \u003cem\u003eObject Detection with Deep Learning Models\u003c/em\u003e [pp. 165\u0026ndash;178]. Chapman and Hall/CRC.\u003c/li\u003e\n\u003cli\u003e[Aditya \u0026amp; Dash, 2022] Aditya, V. K., \u0026amp; Dash, A. K. [2022]. February],\u0026rdquo; Real-life Applications of integration of Augmented Reality and Internet of Things. In \u003cem\u003eSecond International Conference on Artificial Intelligence and Smart Energy [ICAIS]\u003c/em\u003e [pp. 1268\u0026ndash;1273]. IEEE.\u003c/li\u003e\n\u003cli\u003e[Ahmed et al., 2020] Ahmed, H., Younis, E. M., Hendawi, A., \u0026amp; Ali, A. A. [2020]. Heart disease identification from patients\u0026rsquo; social posts, machine learning solution on Spark\u0026rdquo;. \u003cem\u003eFuture Generation Computer Systems\u003c/em\u003e, \u003cem\u003e111\u003c/em\u003e, 714\u0026ndash;722.\u003c/li\u003e\n\u003cli\u003e[F. Ali et al., 2020] Ali, F., El-Sappagh, S., Islam, S. M. R., Kwak, D., Ali, A., Imran, M., \u0026amp; Kwak, K.-S. [2020]. A smart healthcare monitoring system for heart disease prediction based on ensemble deep learning and feature fusion. \u003cem\u003eAn International Journal on Information Fusion\u003c/em\u003e, \u003cem\u003e63\u003c/em\u003e, 208\u0026ndash;222. https://doi.org/10.1016/j.inffus.2020.06.008.\u003c/li\u003e\n\u003cli\u003e[L. Ali et al., 2019] Ali, L., Rahman, A., Khan, A., Zhou, M., Javeed, A., \u0026amp; Khan, J. A. [2019]. An automated diagnostic system for heart disease prediction based on $\\chi^2$ statistical model and optimally configured deep neural network. \u003cem\u003eIEEE Access: Practical Innovations, Open Solutions\u003c/em\u003e, \u003cem\u003e7\u003c/em\u003e, 34938\u0026ndash;34945. https://doi.org/10.1109/access.2019.2904800.\u003c/li\u003e\n\u003cli\u003e[Blanco-Novoa et al., 2020] Blanco-Novoa, \u0026Oacute;., Fraga-Lamas, P., Vilar-Montesinos, A., \u0026amp; Fern\u0026aacute;ndez-Caram\u0026eacute;s, M. [2020]. Creating the internet of augmented things: An open-source framework to make IoT devices and augmented and mixed reality systems talk to each other\u0026rdquo;. \u003cem\u003eSensors\u003c/em\u003e, \u003cem\u003e20\u003c/em\u003e[11].\u003c/li\u003e\n\u003cli\u003e[Dileep et al., 2023] Dileep, P., Rao, K. N., Bodapati, P., Gokuruboyina, S., Peddi, R., Grover, A., \u0026amp; Sheetal, A. [2023]. An automatic heart disease prediction using cluster-based bi-directional LSTM [C-BiLSTM] algorithm. Neural Computing and Applications, 35[10], 7253\u0026ndash;7266.\u003c/li\u003e\n\u003cli\u003e[Elavarasi \u0026amp; Kavitha, 2023] Elavarasi, D., \u0026amp; Kavitha, R. [2023]. December]. Navigating Heart Health with an Elephantine Approach in Clinical Decision Support Systems. In 2nd International Conference on Automation, Computing and Renewable Systems [ICACRS] [pp. 1416\u0026ndash;1423]. IEEE.\u003c/li\u003e\n\u003cli\u003e[Ghorbani, 2024] Ghorbani, F. [2024]. An Intelligent Assistive System Based on Augmented Reality and Internet of Things for Patients with Alzheimer\u0026rsquo;s Disease.\u003c/li\u003e\n\u003cli\u003e[Gomes et al., 2020] Gomes, P., Magaia, N., \u0026amp; Neves, N. [2020]. Industrial and artificial Internet of Things with augmented reality. In Convergence of Artificial Intelligence and the Internet of Things [pp. 323\u0026ndash;346].\u003c/li\u003e\n\u003cli\u003e[Jabeen et al., 2019]Jabeen, F., Maqsood, M., Ghazanfar, M. A., Aadil, F., Khan, S., Khan, M. F., \u0026amp; Mehmood, I. [2019]. An IoT based efficient hybrid recommender system for cardiovascular disease\u0026rdquo;. Peer-to-Peer Networking and Applications, 12, 1263\u0026ndash;1276.\u003c/li\u003e\n\u003cli\u003e[Joko, 2023] Joko, S. [2023]. Development of AR Android Application for Disability and Elderly-Friendly Smart Home Control\u0026rdquo;. International Journal of Information Technology and Computer Engineering [IJITC], 4[02], 32\u0026ndash;42.\u003c/li\u003e\n\u003cli\u003e[Khan, 2020] Khan, M. A. [2020]. An IoT framework for heart disease prediction based on MDCNN classifier. IEEE Access: Practical Innovations, Open Solutions, 8, 34717\u0026ndash;34727. https://doi.org/10.1109/access.2020.2974687.\u003c/li\u003e\n\u003cli\u003e[Kim et al., 2021] Kim, J. C., Laine, T. H., \u0026amp; Ahlund, C. [2021]. Multimodal interaction systems based on internet of things and augmented reality: A systematic literature review\u0026rdquo;. Applied Sciences, 11[4].\u003c/li\u003e\n\u003cli\u003e[Kishore \u0026amp; Jayanthi, 2018] Kishore, A. H., \u0026amp; Jayanthi, V. E. [2018]. Neuro-fuzzy based medical decision support system for coronary artery disease diagnosis and risk level prediction\u0026rdquo;. Journal of Computational and Theoretical Nanoscience, 15[3], 1027\u0026ndash;1037.\u003c/li\u003e\n\u003cli\u003e[Koulouris et al., 2022] Koulouris, D., Menychtas, A., \u0026amp; Maglogiannis, I. [2022]. An IoT-enabled platform for the assessment of physical and mental activities utilizing augmented reality exergaming. Sensors [Basel, Switzerland], 22[9], 3181.https://doi.org/10.3390/s22093181.\u003c/li\u003e\n\u003cli\u003e[Latha \u0026amp; Jeeva, 2019] Latha, C. B. C., \u0026amp; Jeeva, S. C. [2019]. Improving the accuracy of prediction of heart disease risk based on ensemble classification techniques\u0026rdquo;. Informatics in Medicine Unlocked, 16.\u003c/li\u003e\n\u003cli\u003e[Long et al., 2015] Long, N. C., Meesad, P., \u0026amp; Unger, H. [2015]. A highly accurate firefly based algorithm for heart disease prediction\u0026rdquo;. \u003cem\u003eExpert Systems with Applications\u003c/em\u003e, \u003cem\u003e42\u003c/em\u003e[21], 8221\u0026ndash;8231.\u003c/li\u003e\n\u003cli\u003e[Malche et al., 2022] Malche, T., Tharewal, S., Tiwari, P. K., Jabarulla, M. Y., Alnuaim, A. A., Hatamleh, W. A., \u0026amp; Ullah, M. A. [2022] AIoT- based patient activity tracking system for remote patient monitoring. Artificial Intelligence of Things.\u003c/li\u003e\n\u003cli\u003e[Mehmood et al., 2021] Mehmood, A., Iqbal, M., Mehmood, Z., Irtaza, A., Nawaz, M., Nazir, T., \u0026amp; Masood, M. [2021]. Prediction of heart disease using deep convolutional neural networks\u0026rdquo;. \u003cem\u003eArabian Journal for Science and Engineering\u003c/em\u003e, \u003cem\u003e46\u003c/em\u003e[4], 3409\u0026ndash;3422.\u003c/li\u003e\n\u003cli\u003e[Mohan et al., 2019] Mohan, S., Thirumalai, C., \u0026amp; Srivastava, G. [2019]. Effective heart disease prediction using hybrid machine learning techniques. IEEE Access: Practical Innovations, Open Solutions, 7, 81542\u0026ndash;81554. https://doi.org/10.1109/access.2019.2923707.\u003c/li\u003e\n\u003cli\u003e[Morris \u0026amp; Yeboah, 2023] Morris, A., \u0026amp; Yeboah, G. [2023]. Convergence of IoT and Augmented Reality\u0026rdquo;. In Springer Handbook of Augmented Reality [pp. 831\u0026ndash;851]. Springer International Publishing.\u003c/li\u003e\n\u003cli\u003e[Muzammal et al., 2020] Muzammal, M., Talat, R., Sodhro, A. H., \u0026amp; Pirbhulal, S. [2020]. A multi-sensor data fusion enabled ensemble approach for medical data from body sensor networks. An International Journal on Information Fusion, 53, 155\u0026ndash;164. https://doi.org/10.1016/j.inffus.2019.06.021.\u003c/li\u003e\n\u003cli\u003e[Naheem et al., 2023] Naheem, M. R., Aleema, A., Fathima, K., Imran, K., \u0026amp; Fathima, K. [2023] Augmented Reality with Internet of Things. \u003cem\u003e5\u003c/em\u003e, 2582\u0026ndash;5208.\u003c/li\u003e\n\u003cli\u003e[Nancy et al., 2022] Nancy, A. A., Ravindran, D., Vincent, P. D., Srinivasan, K., \u0026amp; Gutierrez Reina, D. [2022]. IoT-cloud-based smart healthcare monitoring system for heart disease prediction via deep learning. Electronics, 11[15].\u003c/li\u003e\n\u003cli\u003e[Patel \u0026amp; Bhatia, 2024] Patel, B., \u0026amp; Bhatia, J. [2024]. A comprehensive review of internet of things and cutting-edge technologies empowering smart farming. Current Science, 2.\u003c/li\u003e\n\u003cli\u003e[Paul et al., 2018] Paul, A. K., Shill, P. C., Rabin, M. R. I., \u0026amp; Murase, K. [2018]. Adaptive weighted fuzzy rule-based system for the risk level assessment of heart disease, Applied Intelligence, 48, 1739\u0026ndash;1756.\u003c/li\u003e\n\u003cli\u003e[Qadri et al., 2020] Qadri, Y. A., Nauman, A., Zikria, Y. B., Vasilakos, A. V., \u0026amp; Kim, S. W. [2020]. The future of healthcare internet of things: a survey of emerging technologies. IEEE Communications Surveys \u0026amp; Tutorials, 22[2], 1121\u0026ndash;1167.\u003c/li\u003e\n\u003cli\u003e[S. Razdan et al., 2022] Razdan, S., Sharma, S., Manickam, P., Mariappan, S. A., Murugesan, S. M., Hansda, S., Kaushik, A., Shinde, R., \u0026amp; Thipperudraswamy, S. P. [2022]. Artificial intelligence [AI] and internet of medical things [IoMT] assisted biomedical systems for intelligent healthcare. \u003cem\u003eBiosensors\u003c/em\u003e, \u003cem\u003e12\u003c/em\u003e[8].\u003c/li\u003e\n\u003cli\u003e[Sahshanu Razdan \u0026amp; Sharma, 2022] Razdan, Sahshanu, \u0026amp; Sharma, S. [2022]. Internet of medical things [IoMT]: Overview, emerging technologies, and case studies. \u003cem\u003eIETE Technical Review\u003c/em\u003e, \u003cem\u003e39\u003c/em\u003e[4], 775\u0026ndash;788. https://doi.org/10.1080/02564602.2021.1927863.\u003c/li\u003e\n\u003cli\u003e[Sahu et al., 2024] Sahu, M., Gupta, R., Ambasta, R. K., \u0026amp; Kumar, P. [2024]. IoT-driven Augmented Reality and Virtual Reality Systems in Neurological Sciences, Internet of Things, 101098.\u003c/li\u003e\n\u003cli\u003e[Samuel et al., 2017] Samuel, O. W., Asogbon, G. M., Sangaiah, A. K., Fang, P., \u0026amp; Li, G. [2017]. An integrated decision support system based on ANN and Fuzzy AHP for heart failure risk prediction. \u003cem\u003eExpert Systems with Applications\u003c/em\u003e, \u003cem\u003e68\u003c/em\u003e, 163\u0026ndash;172.\u003c/li\u003e\n\u003cli\u003e[Sassi \u0026amp; Fourati, 2020] Sassi, H., \u0026amp; Fourati, M. S. [2020]. Architecture for visualizing indoor air quality data with augmented reality based cognitive Internet of Things. In Advanced Information Networking and Applications: Proceedings of the 34th International Conference on Advanced Information Networking and Applications [AINA-2020] [pp. 405\u0026ndash;418]. Springer International Publishing.\u003c/li\u003e\n\u003cli\u003e[Shi et al., 2020] Shi, Q., Dong, B., He, T., Sun, Z., Zhu, J., Zhang, Z., \u0026amp; Lee, C. [2020]. Progress in wearable electronics/photonics\u0026mdash;Moving toward the era of artificial intelligence and internet of things. InfoMat, 2[6], 1131\u0026ndash;1162. https://doi.org/10.1002/inf2.12122.\u003c/li\u003e\n\u003cli\u003e[Siang et al., 2023] Siang, H. T., Gobee, S., \u0026amp; Durairajah, V. [2023]. Development of augmented reality control and monitoring for smart manufacturing plant with IoT interface. In AIP Conference Proceedings [Vol. 2788]. AIP Publishing.\u003c/li\u003e\n\u003cli\u003e[Van Pham et al., 2018] Van Pham, H., Son, L. H., \u0026amp; Tuan, L. M. [2018]. A proposal of expert system using deep learning neural networks and fuzzy rules for diagnosing heart disease. In Frontiers in Intelligent Computing: Theory and Applications: Proceedings of the 7th International Conference on FICTA [Vol. 1, pp. 189\u0026ndash;198]. Springer.\u003c/li\u003e\n\u003cli\u003e[Zhang et al., 2021] Zhang, D., Chen, Y., Chen, Y., Ye, S., Cai, W., Jiang, J., \u0026amp; Chen, M. [2021]. Heart disease prediction based on the embedded feature selection method and deep neural network\u0026rdquo;. Journal of Healthcare Engineering, 2021, 1\u0026ndash;9.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"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":"IoT, Augmented reality, Expectation Maximization based clustering, Tracking employee’s healthcare status, Intelligent swarm dependent BAT optimization approach, Hybrid Inception V3 and MobileNet V2 model.","lastPublishedDoi":"10.21203/rs.3.rs-5244034/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5244034/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThe growth of IoT in modern days helps devices to work productively and efficiently. However, the device needs human intervention for timely assistance. The proposed design consists of transmitter and receiver side, the transmitter side comprises of various sensors connected with the Raspberry Pi module and is connected to the camera and Wi-Fi module. The transmitted data are recognized and classified at the receiver side for any unstable healthcare condition of the employees or patients which could be connected and monitored over Augmented reality. Once the abnormal status is detected, an alert will be sent to the caretaker or physician to take necessary actions. The performance estimation is made in terms of various performance metrics and the comparison made with the existing system reveals that the proposed model is effective over other conventional schemes.\u003c/p\u003e","manuscriptTitle":"A Novel Healthcare Monitoring System using Optimal Hybrid DL, AR and IoT","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-12-17 06:34:19","doi":"10.21203/rs.3.rs-5244034/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":"df55fef2-51fa-4537-a904-523a3ead8aaa","owner":[],"postedDate":"December 17th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-02-11T14:23:48+00:00","versionOfRecord":[],"versionCreatedAt":"2024-12-17 06:34:19","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-5244034","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5244034","identity":"rs-5244034","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}