Pub Date : 2022-03-09DOI: 10.1109/ESCI53509.2022.9758349
Priyanka Prabhakaran, S. Anandakumar, E. Priyanka
Railways have been expanding its roots since their inception from conventional ballasted rail systems to ballastless metro's and high-speed rail. Metro rail systems are predicted to revolutionise the transportation sector's outlook practically in every metropolitan city catering to the needs of day-to-day routine traffic essence. In order to provide fast and efficient transportation modes the railways are dependent on moving and non-moving components. One among the major non movable component is said to be the railway tracks commonly known as railroad systems. Railroad systems are prone to regular maintenance interventions depending on traffic intensity and various other external factors namely rail temperature, climatic variations etc. Periodic and corrective maintenance activities are disrupted by service runs during daytime and hence they are planned to be performed overnight. In order to perform effective railroad maintenance a proper schedule is required along with the intervention requirement rate. The study adopts clustering algorithm to identify the probability of intervention rates by categorizing the maintenance interventions into three probability levels namely low, medium, and high. The results of the study indicate that the segments falling in the category of medium levels require higher maintenance intervention than the high and low severity levels.
{"title":"Railroad Maintenance Predictor System for Metro Railroad Systems","authors":"Priyanka Prabhakaran, S. Anandakumar, E. Priyanka","doi":"10.1109/ESCI53509.2022.9758349","DOIUrl":"https://doi.org/10.1109/ESCI53509.2022.9758349","url":null,"abstract":"Railways have been expanding its roots since their inception from conventional ballasted rail systems to ballastless metro's and high-speed rail. Metro rail systems are predicted to revolutionise the transportation sector's outlook practically in every metropolitan city catering to the needs of day-to-day routine traffic essence. In order to provide fast and efficient transportation modes the railways are dependent on moving and non-moving components. One among the major non movable component is said to be the railway tracks commonly known as railroad systems. Railroad systems are prone to regular maintenance interventions depending on traffic intensity and various other external factors namely rail temperature, climatic variations etc. Periodic and corrective maintenance activities are disrupted by service runs during daytime and hence they are planned to be performed overnight. In order to perform effective railroad maintenance a proper schedule is required along with the intervention requirement rate. The study adopts clustering algorithm to identify the probability of intervention rates by categorizing the maintenance interventions into three probability levels namely low, medium, and high. The results of the study indicate that the segments falling in the category of medium levels require higher maintenance intervention than the high and low severity levels.","PeriodicalId":436539,"journal":{"name":"2022 International Conference on Emerging Smart Computing and Informatics (ESCI)","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127982919","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-03-09DOI: 10.1109/ESCI53509.2022.9758273
S. Rajgade, P. Shendge
This paper presents, Equivalent Input Disturbance (EID) based robust control for Robot Manipulator (RM) under uncertain conditions. Robust control is achieved by designing the Generalised Extended State Observer (GESO). Modern control designs typically requires a complete state vector for their implementations as well as estimation of uncertain states. However above mentioned requirement are difficult to meet in a real life system. To fulfil aforementioned requirements a GESO is developed that simultaneously estimates the state vector and the uncertainty. To begin the nonlinear dynamics of a robot manipulator are modelled using a linear formulation that includes uncertainty and disturbances. Then for disturbance rejection an EID + GESO based control is presented. Uncertainty and disturbances are addressed as a single lump disturbance called EID which is effectively attenuated using GESO and state feedback. The approach closed loop stability is also proven. The method is then deployed to a robot manipulator and its usefulness is demonstrated using numerical simulations under significant uncertainty and disturbances to illustrate the efficacy and robustness of the suggested design.
{"title":"Equivalent Input Disturbance Based Robust Control For Robot Manipulator Using Generalised Extended State Observer","authors":"S. Rajgade, P. Shendge","doi":"10.1109/ESCI53509.2022.9758273","DOIUrl":"https://doi.org/10.1109/ESCI53509.2022.9758273","url":null,"abstract":"This paper presents, Equivalent Input Disturbance (EID) based robust control for Robot Manipulator (RM) under uncertain conditions. Robust control is achieved by designing the Generalised Extended State Observer (GESO). Modern control designs typically requires a complete state vector for their implementations as well as estimation of uncertain states. However above mentioned requirement are difficult to meet in a real life system. To fulfil aforementioned requirements a GESO is developed that simultaneously estimates the state vector and the uncertainty. To begin the nonlinear dynamics of a robot manipulator are modelled using a linear formulation that includes uncertainty and disturbances. Then for disturbance rejection an EID + GESO based control is presented. Uncertainty and disturbances are addressed as a single lump disturbance called EID which is effectively attenuated using GESO and state feedback. The approach closed loop stability is also proven. The method is then deployed to a robot manipulator and its usefulness is demonstrated using numerical simulations under significant uncertainty and disturbances to illustrate the efficacy and robustness of the suggested design.","PeriodicalId":436539,"journal":{"name":"2022 International Conference on Emerging Smart Computing and Informatics (ESCI)","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132944598","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-03-09DOI: 10.1109/ESCI53509.2022.9758356
Pinakshi Panda, Ankur Priyadarshi
Cancer now a day is playing a vital role in increasing the number of deaths throughout the world. Early detection of cancer increases the degree of recovery. Machine Learning has given various models based on biopsy data and the microarray data for cancer classification. The microarray data is having high dimension. Hence applying machine learning algorithm is directly applied to the microarray data for classification purposes then it will face the Small Sample Size (SSS) problem. So, before classification, the dimension of the dataset has to be reduced by using any available technique. In this research work an integrated approach based on the RFE-ACO-RF method has been proposed as a cancer diagnosis model. The RFE will be used for feature selection purpose, ACO is used for optimization purpose and the RF for classification purpose. The performance of the model will be calculated based on accuracy, F1 score, precision and recall.
{"title":"RFE-ACO-RF: An approach for Cancer Microarray Data Diagnosis","authors":"Pinakshi Panda, Ankur Priyadarshi","doi":"10.1109/ESCI53509.2022.9758356","DOIUrl":"https://doi.org/10.1109/ESCI53509.2022.9758356","url":null,"abstract":"Cancer now a day is playing a vital role in increasing the number of deaths throughout the world. Early detection of cancer increases the degree of recovery. Machine Learning has given various models based on biopsy data and the microarray data for cancer classification. The microarray data is having high dimension. Hence applying machine learning algorithm is directly applied to the microarray data for classification purposes then it will face the Small Sample Size (SSS) problem. So, before classification, the dimension of the dataset has to be reduced by using any available technique. In this research work an integrated approach based on the RFE-ACO-RF method has been proposed as a cancer diagnosis model. The RFE will be used for feature selection purpose, ACO is used for optimization purpose and the RF for classification purpose. The performance of the model will be calculated based on accuracy, F1 score, precision and recall.","PeriodicalId":436539,"journal":{"name":"2022 International Conference on Emerging Smart Computing and Informatics (ESCI)","volume":"206 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131978409","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-03-09DOI: 10.1109/ESCI53509.2022.9758270
Sofiya Mujawar, Jaya Gupta
Medical disease detection is a vast field of image, signal and video processing that involves a large number of complex operations, which include but are not limited to data acquisition, pre-processing, segmentation, feature extraction, feature selection, classification and post-processing. The efficiency of signal classification is directly proportional to the efficiency with which these internal blocks are designed. In order to improve the efficiency of these blocks, several bio-inspired optimization algorithms are proposed by researchers. These include but are not limited to, Particle Swarm Optimization (PSO), Genetic Algorithm (GA), Neural Networks (NN), etc. Each of these algorithms can be applied to optimize individual signal processing blocks, thereby improving overall system performance. Due to a large variety of available bio-inspired algorithms, it is ambiguous for system designers to select the best possible algorithmic combination for their medical disease classification design. In order to reduce this ambiguity, the underlying text evaluates performance of some of the most efficient bio-inspired algorithms, and statistically compares them on basis of their application. These applications vary w.r.t. identified disease, type of signal being processed, etc. This comparison will assist researchers and system designers to develop highly efficient medical disease classification systems for clinical use.
{"title":"A Statistical Perspective for Empirical Analysis of Bio-Inspired Algorithms for Medical Disease Detection","authors":"Sofiya Mujawar, Jaya Gupta","doi":"10.1109/ESCI53509.2022.9758270","DOIUrl":"https://doi.org/10.1109/ESCI53509.2022.9758270","url":null,"abstract":"Medical disease detection is a vast field of image, signal and video processing that involves a large number of complex operations, which include but are not limited to data acquisition, pre-processing, segmentation, feature extraction, feature selection, classification and post-processing. The efficiency of signal classification is directly proportional to the efficiency with which these internal blocks are designed. In order to improve the efficiency of these blocks, several bio-inspired optimization algorithms are proposed by researchers. These include but are not limited to, Particle Swarm Optimization (PSO), Genetic Algorithm (GA), Neural Networks (NN), etc. Each of these algorithms can be applied to optimize individual signal processing blocks, thereby improving overall system performance. Due to a large variety of available bio-inspired algorithms, it is ambiguous for system designers to select the best possible algorithmic combination for their medical disease classification design. In order to reduce this ambiguity, the underlying text evaluates performance of some of the most efficient bio-inspired algorithms, and statistically compares them on basis of their application. These applications vary w.r.t. identified disease, type of signal being processed, etc. This comparison will assist researchers and system designers to develop highly efficient medical disease classification systems for clinical use.","PeriodicalId":436539,"journal":{"name":"2022 International Conference on Emerging Smart Computing and Informatics (ESCI)","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127816689","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-03-09DOI: 10.1109/ESCI53509.2022.9758227
I. Kansal, Renu Popli, Jyoti Verma, Vivek Bhardwaj, R. Bhardwaj
Health care and well-being are concerned with the upkeep or maintenance of humans through preventative medicine, diagnosis, therapies, regeneration, or prevention of disease, ailment, injury, and other health-related conditions in people. Healthcare is unique in comparison to other industries. It is an elevated segment, and people expect the best possible care and services at all costs. Through continuous integration and resource optimization, the use of IoT technology in health applications enables the health care industry to improve care quality while lowering costs. The IoT in diagnostic imaging enables real-time identification and correction of imaging apparatus parameters due to the ease with which imaging apparatus parameters can be auto-analyzed. This paper discusses the impact of online image processing methods in IoT-based health care, which can be beneficial in the health sector for predicting some major human diseases. Due to individuality, image complex nature, extensive variation between interpreters, and fatigue, human experts' ability to interpret images is quite limited. We focus on the role of Digital Image Processing in disease detection, Image Dataset Preparation for Machine and Deep Learning, the role of Digital Image Processing in IOT based applications of health care, a case study of IoT-based healthcare application of disease classification.
{"title":"Digital Image Processing and IoT in Smart Health Care -A review","authors":"I. Kansal, Renu Popli, Jyoti Verma, Vivek Bhardwaj, R. Bhardwaj","doi":"10.1109/ESCI53509.2022.9758227","DOIUrl":"https://doi.org/10.1109/ESCI53509.2022.9758227","url":null,"abstract":"Health care and well-being are concerned with the upkeep or maintenance of humans through preventative medicine, diagnosis, therapies, regeneration, or prevention of disease, ailment, injury, and other health-related conditions in people. Healthcare is unique in comparison to other industries. It is an elevated segment, and people expect the best possible care and services at all costs. Through continuous integration and resource optimization, the use of IoT technology in health applications enables the health care industry to improve care quality while lowering costs. The IoT in diagnostic imaging enables real-time identification and correction of imaging apparatus parameters due to the ease with which imaging apparatus parameters can be auto-analyzed. This paper discusses the impact of online image processing methods in IoT-based health care, which can be beneficial in the health sector for predicting some major human diseases. Due to individuality, image complex nature, extensive variation between interpreters, and fatigue, human experts' ability to interpret images is quite limited. We focus on the role of Digital Image Processing in disease detection, Image Dataset Preparation for Machine and Deep Learning, the role of Digital Image Processing in IOT based applications of health care, a case study of IoT-based healthcare application of disease classification.","PeriodicalId":436539,"journal":{"name":"2022 International Conference on Emerging Smart Computing and Informatics (ESCI)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129880734","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-03-09DOI: 10.1109/ESCI53509.2022.9758234
A. Shukla, Nitin Lodha
Demand of block-chain is increasing day by day. On other hand smart contract are frequently used in commercial application. Artificial intelligence mechanism is capability to making things intelligent. Proposed paper is investigation role of artificial intelligence in development of smart contract that are block chain based. Deep learning approach is the feature of Artificial intelligence that could make the smart contract running on block chain more efficient and smart. Several existing researches that are made in area of block chain and AI have been considered in present research. The issues faced in previous research are their limited scope and lack of accuracy and performance. The proposed work is supposed to provide better solution in term of accuracy and performance.
{"title":"Investigating the Role of Artificial Intelligence in Building Smart Contact on Block-Chain","authors":"A. Shukla, Nitin Lodha","doi":"10.1109/ESCI53509.2022.9758234","DOIUrl":"https://doi.org/10.1109/ESCI53509.2022.9758234","url":null,"abstract":"Demand of block-chain is increasing day by day. On other hand smart contract are frequently used in commercial application. Artificial intelligence mechanism is capability to making things intelligent. Proposed paper is investigation role of artificial intelligence in development of smart contract that are block chain based. Deep learning approach is the feature of Artificial intelligence that could make the smart contract running on block chain more efficient and smart. Several existing researches that are made in area of block chain and AI have been considered in present research. The issues faced in previous research are their limited scope and lack of accuracy and performance. The proposed work is supposed to provide better solution in term of accuracy and performance.","PeriodicalId":436539,"journal":{"name":"2022 International Conference on Emerging Smart Computing and Informatics (ESCI)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130979989","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-03-09DOI: 10.1109/ESCI53509.2022.9758348
P. Rajesh, D. Akila
E-learning has piqued the interest of companies, educational institutions, and people alike. E-learning systems are becoming increasingly prominent as an educational trend. It typically refers to educational attempts spread via the use of computers in an attempt to transmit information. Students can engage with other students and discuss questions about certain topics thanks to e-Learning platforms and similar technologies. Teachers, on the other hand, frequently remain outside of this process and are unaware of the learning issues that exist in their classes. Adopting a Sentiment Analysis approach for detecting the student mood throughout the learning process might be a solution for better learning method. In this paper, we used sentimental analysis on E-learning data. SVM and Naïve Bayes algorithms are fused to be used as a Hybrid algorithm for better accuracy. Performance analysis shows that state-of-art methods like Naïve Bayes and SVM algorithms give 90% and 94% respectively whereas our proposed hybrid method gives approximately 97% of accuracy.
{"title":"Sentimental analysis on E-Learning videos using Hybrid Algorithm based on Naïve Bayes and SVM","authors":"P. Rajesh, D. Akila","doi":"10.1109/ESCI53509.2022.9758348","DOIUrl":"https://doi.org/10.1109/ESCI53509.2022.9758348","url":null,"abstract":"E-learning has piqued the interest of companies, educational institutions, and people alike. E-learning systems are becoming increasingly prominent as an educational trend. It typically refers to educational attempts spread via the use of computers in an attempt to transmit information. Students can engage with other students and discuss questions about certain topics thanks to e-Learning platforms and similar technologies. Teachers, on the other hand, frequently remain outside of this process and are unaware of the learning issues that exist in their classes. Adopting a Sentiment Analysis approach for detecting the student mood throughout the learning process might be a solution for better learning method. In this paper, we used sentimental analysis on E-learning data. SVM and Naïve Bayes algorithms are fused to be used as a Hybrid algorithm for better accuracy. Performance analysis shows that state-of-art methods like Naïve Bayes and SVM algorithms give 90% and 94% respectively whereas our proposed hybrid method gives approximately 97% of accuracy.","PeriodicalId":436539,"journal":{"name":"2022 International Conference on Emerging Smart Computing and Informatics (ESCI)","volume":"120 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133975113","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-03-09DOI: 10.1109/ESCI53509.2022.9758303
Hari Chandana B, Divya Nune, Harika D
The Circular patch antenna with a C Slot is developed to operate at 4.95 GHz using textile material. These structures are developed using textile materials like Zelt as radiating element and Felt as substrate material. The results of simulations are observed to give better performance with arrays rather than a single antenna in HFSS Software tool. The developed antennas report good performance when compared to existing antennas. The parameters used to compare the two antennas are Gain, VSWR, Return Loss, Efficiency, etc to suit human body. The gain and efficiency for 2 element array were improved by 37.8% and 7.8% respectively over that of single antenna with a tradeoff in bandwidth by 22.7%.
{"title":"Design and Simulation of Circular Textile Antenna with C- Slot","authors":"Hari Chandana B, Divya Nune, Harika D","doi":"10.1109/ESCI53509.2022.9758303","DOIUrl":"https://doi.org/10.1109/ESCI53509.2022.9758303","url":null,"abstract":"The Circular patch antenna with a C Slot is developed to operate at 4.95 GHz using textile material. These structures are developed using textile materials like Zelt as radiating element and Felt as substrate material. The results of simulations are observed to give better performance with arrays rather than a single antenna in HFSS Software tool. The developed antennas report good performance when compared to existing antennas. The parameters used to compare the two antennas are Gain, VSWR, Return Loss, Efficiency, etc to suit human body. The gain and efficiency for 2 element array were improved by 37.8% and 7.8% respectively over that of single antenna with a tradeoff in bandwidth by 22.7%.","PeriodicalId":436539,"journal":{"name":"2022 International Conference on Emerging Smart Computing and Informatics (ESCI)","volume":"111 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133048273","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-03-09DOI: 10.1109/ESCI53509.2022.9758235
A. Rajasekaran, D. Yuvaraj, Azees Maria
In the medical business, the Internet of Medical Things (IoMT) has turned into a specialist application framework. It is utilized to accumulate and break down the physiological boundaries of patients. The vital body parameters are analyzed by the medical sensor-hubs, which are embedded in the patient's body. It would thusly detect the patient's medical data by utilizing convenient gadgets. Since patient data is so delicate to uncover without the help of a medical expert, the security and protection of medical information are turning into a difficult issue for the IoMT. Hence, an anonymous authentication protocol based on IoT is likely to be presented in this paper. This work is liked to determine the security protection issues in the IoMT. In this work, a secure and anonymous patient and medical advisor authentication scheme is proposed to guarantee secure correspondence in medical care applications. The validation of the work is evaluated in terms of computational cost with different existing schemes using the Cygwin platform.
{"title":"Secure Authentication Scheme for Medical care applications based on IoT","authors":"A. Rajasekaran, D. Yuvaraj, Azees Maria","doi":"10.1109/ESCI53509.2022.9758235","DOIUrl":"https://doi.org/10.1109/ESCI53509.2022.9758235","url":null,"abstract":"In the medical business, the Internet of Medical Things (IoMT) has turned into a specialist application framework. It is utilized to accumulate and break down the physiological boundaries of patients. The vital body parameters are analyzed by the medical sensor-hubs, which are embedded in the patient's body. It would thusly detect the patient's medical data by utilizing convenient gadgets. Since patient data is so delicate to uncover without the help of a medical expert, the security and protection of medical information are turning into a difficult issue for the IoMT. Hence, an anonymous authentication protocol based on IoT is likely to be presented in this paper. This work is liked to determine the security protection issues in the IoMT. In this work, a secure and anonymous patient and medical advisor authentication scheme is proposed to guarantee secure correspondence in medical care applications. The validation of the work is evaluated in terms of computational cost with different existing schemes using the Cygwin platform.","PeriodicalId":436539,"journal":{"name":"2022 International Conference on Emerging Smart Computing and Informatics (ESCI)","volume":"45 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128194379","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-03-09DOI: 10.1109/ESCI53509.2022.9758337
Renuka D. Suryawanshi, S. Vanjale, M. Vanjale
Electroencephalography (EEG) signals are a combination of complex pattern sequences, which are periodic in nature. These pattern sequences include a gamma waves that indicates deep thinking behaviour, a beta wave sequence that indicates busy and active mind status, an alpha wave segment which indicates reflective and restful behaviour, a theta wave which is an indicative of drowsiness, and a delta wave which indicates sleeping & dreaming conditions. Features like frequency changes, amplitude changes, pattern changes, etc. are used to identify chronic, ischemic and other diseases related to the brain. In order to classify these wave patterns into brain diseases like epilepsy, a series of high complexity signal processing operations are needed to be executed in tandem. These operations include signal pre-processing, feature extraction, feature selection, classification into epileptic & non-epileptic seizure and post-processing. A large variety of algorithms are developed by researchers for each of these operations. Performance of these algorithms varies largely w.r.t. the number of leads used for EEG capture, filtering efficiency, feature extraction & selection efficiency, and classifier efficiency. Thus, it becomes ambiguous for researchers and system designers to select the best possible algorithm set for their application. In order to reduce the ambiguity, this text provides a comprehensive comparison of a wide variety of epileptic & non-epileptic seizure classification system models. These models are statistically compared on the basis of overall accuracy, delay of decision making, precision, recall, f-measure and field of application. It is observed that convolutional neural network (CNN) based models outperform other models in terms of general-purpose performance, while specialized CNN models must be used for application specific deployments.
{"title":"A Fuzzy Statistical Perspective for Empirical Evaluation of EEG Classification Models for Epileptic Seizures","authors":"Renuka D. Suryawanshi, S. Vanjale, M. Vanjale","doi":"10.1109/ESCI53509.2022.9758337","DOIUrl":"https://doi.org/10.1109/ESCI53509.2022.9758337","url":null,"abstract":"Electroencephalography (EEG) signals are a combination of complex pattern sequences, which are periodic in nature. These pattern sequences include a gamma waves that indicates deep thinking behaviour, a beta wave sequence that indicates busy and active mind status, an alpha wave segment which indicates reflective and restful behaviour, a theta wave which is an indicative of drowsiness, and a delta wave which indicates sleeping & dreaming conditions. Features like frequency changes, amplitude changes, pattern changes, etc. are used to identify chronic, ischemic and other diseases related to the brain. In order to classify these wave patterns into brain diseases like epilepsy, a series of high complexity signal processing operations are needed to be executed in tandem. These operations include signal pre-processing, feature extraction, feature selection, classification into epileptic & non-epileptic seizure and post-processing. A large variety of algorithms are developed by researchers for each of these operations. Performance of these algorithms varies largely w.r.t. the number of leads used for EEG capture, filtering efficiency, feature extraction & selection efficiency, and classifier efficiency. Thus, it becomes ambiguous for researchers and system designers to select the best possible algorithm set for their application. In order to reduce the ambiguity, this text provides a comprehensive comparison of a wide variety of epileptic & non-epileptic seizure classification system models. These models are statistically compared on the basis of overall accuracy, delay of decision making, precision, recall, f-measure and field of application. It is observed that convolutional neural network (CNN) based models outperform other models in terms of general-purpose performance, while specialized CNN models must be used for application specific deployments.","PeriodicalId":436539,"journal":{"name":"2022 International Conference on Emerging Smart Computing and Informatics (ESCI)","volume":"40 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127564749","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：