TY - GEN
T1 - EEG-based Anxiety Detection with Feature Selection
AU - Park, Jeongyeong
AU - Jin, Nanlin
AU - Chen, Jianjun
AU - Qi, Jun
N1 - Publisher Copyright:
© 2023 IEEE.
PY - 2023
Y1 - 2023
N2 - Anxiety has become a major public health problem, and accurate indicators of diagnosis are required. Traditional assessments to identify anxiety are quantitative questionnaires, which lack objective criteria that incorporate physiological changes in the body. In recent years, electroencephalography (EEG) has been demonstrated to be a convenient and cost-effective supplementary diagnostic tool. This study presents the methods for selecting the features to extract EEG characteristics. We also employ statistical analyses on feature selection to extract in total 41 important features to objective identify anxiety. The classification results of using these selected and created features were verified through public dataset DASPS. Random Forest has been found to have shown the best performance with a ROC AUC value of 0.93, a recall value of 0.931, and a precision of 94.7%. These findings suggest that EEG signals can potentially serve as reliable indicators to detect anxiety, while providing explainable characteristics that can associate anxiety.
AB - Anxiety has become a major public health problem, and accurate indicators of diagnosis are required. Traditional assessments to identify anxiety are quantitative questionnaires, which lack objective criteria that incorporate physiological changes in the body. In recent years, electroencephalography (EEG) has been demonstrated to be a convenient and cost-effective supplementary diagnostic tool. This study presents the methods for selecting the features to extract EEG characteristics. We also employ statistical analyses on feature selection to extract in total 41 important features to objective identify anxiety. The classification results of using these selected and created features were verified through public dataset DASPS. Random Forest has been found to have shown the best performance with a ROC AUC value of 0.93, a recall value of 0.931, and a precision of 94.7%. These findings suggest that EEG signals can potentially serve as reliable indicators to detect anxiety, while providing explainable characteristics that can associate anxiety.
UR - http://www.scopus.com/inward/record.url?scp=85186761965&partnerID=8YFLogxK
U2 - 10.1109/CyberC58899.2023.00016
DO - 10.1109/CyberC58899.2023.00016
M3 - Conference Proceeding
AN - SCOPUS:85186761965
T3 - Proceedings - 2023 International Conference on Cyber-Enabled Distributed Computing and Knowledge Discovery, CyberC 2023
SP - 32
EP - 39
BT - Proceedings - 2023 International Conference on Cyber-Enabled Distributed Computing and Knowledge Discovery, CyberC 2023
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 15th International Conference on Cyber-Enabled Distributed Computing and Knowledge Discovery, CyberC 2023
Y2 - 2 November 2023 through 4 November 2023
ER -
