Himanshu Chhabra, Diksha Sharma, Urvashi Chauhan, Lakhan Dev Sharma
{"title":"基于脑电信号和特定导联提升小波变换的认知负荷检测。","authors":"Himanshu Chhabra, Diksha Sharma, Urvashi Chauhan, Lakhan Dev Sharma","doi":"10.1088/2057-1976/ade15a","DOIUrl":null,"url":null,"abstract":"<p><p>Solving an arithmetic task is a complex assignment that includes sequencing, memory, fact retrieval, and decision making. Observation of the human brain's response to such activities is quite essential as it helps in the diagnosis of various diseases and also facilitates understanding the brain's response under stressful conditions. Thus, the present work focuses on the development of an effective approach for recognizing mental arithmetic load by the analysis of electroencephalographic (EEG) signals with the advancement of lead minimization. The proposed technique uses only two frontal lead (Fp1 and Fp2) EEG data in order to present a cost-effective technique with reduced complexity. By applying the lifting wavelet processing approach, the EEG signal is divided into 12 different frequency band segments.Furthermore, fuzzy entropy features were obtained and, to choose the 100 most important features for additional processing, the lowest redundancy maximum relevance technique was used. In this study, three supervised machine learning models were successfully used: closest neighbor (KNN), support vector machine (SVM) and random forest algorithm (RFA) to categorize EEG data while performing and resting states of a mental arithmetic task. The classification accuracy has been calculated in two cases, that is, (i) when EEG data extracted from 19 leads is utilized and (ii) when EEG data extracted from 2 frontal leads are utilized. In case (i) , the SVM classifier gives the best accuracy among all three classifiers, that is, 96. 63 %. In addition, for the classification accuracy with specific lead selection (Fp1 and Fp2), the SVM classifier provides the highest accuracy of 95.34 %. Thus, the proposed technique gives preferable classification accuracy with reduced number of EEG leads. Thus, this technique is suitable for designing wearable devices for cognitive load detection.
.</p>","PeriodicalId":8896,"journal":{"name":"Biomedical Physics & Engineering Express","volume":" ","pages":""},"PeriodicalIF":1.3000,"publicationDate":"2025-06-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Detection of Cognitive Load using EEG Signal and Lifting Wavelet Transform with Specific Lead Selection.\",\"authors\":\"Himanshu Chhabra, Diksha Sharma, Urvashi Chauhan, Lakhan Dev Sharma\",\"doi\":\"10.1088/2057-1976/ade15a\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p><p>Solving an arithmetic task is a complex assignment that includes sequencing, memory, fact retrieval, and decision making. Observation of the human brain's response to such activities is quite essential as it helps in the diagnosis of various diseases and also facilitates understanding the brain's response under stressful conditions. Thus, the present work focuses on the development of an effective approach for recognizing mental arithmetic load by the analysis of electroencephalographic (EEG) signals with the advancement of lead minimization. The proposed technique uses only two frontal lead (Fp1 and Fp2) EEG data in order to present a cost-effective technique with reduced complexity. By applying the lifting wavelet processing approach, the EEG signal is divided into 12 different frequency band segments.Furthermore, fuzzy entropy features were obtained and, to choose the 100 most important features for additional processing, the lowest redundancy maximum relevance technique was used. In this study, three supervised machine learning models were successfully used: closest neighbor (KNN), support vector machine (SVM) and random forest algorithm (RFA) to categorize EEG data while performing and resting states of a mental arithmetic task. The classification accuracy has been calculated in two cases, that is, (i) when EEG data extracted from 19 leads is utilized and (ii) when EEG data extracted from 2 frontal leads are utilized. In case (i) , the SVM classifier gives the best accuracy among all three classifiers, that is, 96. 63 %. In addition, for the classification accuracy with specific lead selection (Fp1 and Fp2), the SVM classifier provides the highest accuracy of 95.34 %. Thus, the proposed technique gives preferable classification accuracy with reduced number of EEG leads. Thus, this technique is suitable for designing wearable devices for cognitive load detection.
.</p>\",\"PeriodicalId\":8896,\"journal\":{\"name\":\"Biomedical Physics & Engineering Express\",\"volume\":\" \",\"pages\":\"\"},\"PeriodicalIF\":1.3000,\"publicationDate\":\"2025-06-05\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Biomedical Physics & Engineering Express\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1088/2057-1976/ade15a\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q3\",\"JCRName\":\"RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Biomedical Physics & Engineering Express","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1088/2057-1976/ade15a","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING","Score":null,"Total":0}
Detection of Cognitive Load using EEG Signal and Lifting Wavelet Transform with Specific Lead Selection.
Solving an arithmetic task is a complex assignment that includes sequencing, memory, fact retrieval, and decision making. Observation of the human brain's response to such activities is quite essential as it helps in the diagnosis of various diseases and also facilitates understanding the brain's response under stressful conditions. Thus, the present work focuses on the development of an effective approach for recognizing mental arithmetic load by the analysis of electroencephalographic (EEG) signals with the advancement of lead minimization. The proposed technique uses only two frontal lead (Fp1 and Fp2) EEG data in order to present a cost-effective technique with reduced complexity. By applying the lifting wavelet processing approach, the EEG signal is divided into 12 different frequency band segments.Furthermore, fuzzy entropy features were obtained and, to choose the 100 most important features for additional processing, the lowest redundancy maximum relevance technique was used. In this study, three supervised machine learning models were successfully used: closest neighbor (KNN), support vector machine (SVM) and random forest algorithm (RFA) to categorize EEG data while performing and resting states of a mental arithmetic task. The classification accuracy has been calculated in two cases, that is, (i) when EEG data extracted from 19 leads is utilized and (ii) when EEG data extracted from 2 frontal leads are utilized. In case (i) , the SVM classifier gives the best accuracy among all three classifiers, that is, 96. 63 %. In addition, for the classification accuracy with specific lead selection (Fp1 and Fp2), the SVM classifier provides the highest accuracy of 95.34 %. Thus, the proposed technique gives preferable classification accuracy with reduced number of EEG leads. Thus, this technique is suitable for designing wearable devices for cognitive load detection.
.
期刊介绍:
BPEX is an inclusive, international, multidisciplinary journal devoted to publishing new research on any application of physics and/or engineering in medicine and/or biology. Characterized by a broad geographical coverage and a fast-track peer-review process, relevant topics include all aspects of biophysics, medical physics and biomedical engineering. Papers that are almost entirely clinical or biological in their focus are not suitable. The journal has an emphasis on publishing interdisciplinary work and bringing research fields together, encompassing experimental, theoretical and computational work.