TY - GEN
T1 - A Comprehensive Analysis of Alcoholic EEG Signals with Detrend Fluctuation Analysis and Post Classifiers
AU - Prabhakar, Sunil Kumar
AU - Rajaguru, Harikumar
AU - Lee, Seong Whan
N1 - Funding Information:
During the consumption of alcohol by a human being, the activities of the brain changes from the normal This work was supported by the Institute for Information & Communications Technology Planning & Evaluation (IITP) grant funded by the Korea government (No. 2017-0-00451; Development of BCI based Brain and Cognitive Computing Technology for Recognizing User ’ s Intentions using Deep Learning).
Publisher Copyright:
© 2019 IEEE.
PY - 2019/2
Y1 - 2019/2
N2 - Different pathological and physiological activities of the brain can be analyzed by means of utilizing Electroencephalography (EEG) signals. One such important activity which can be assessed and understood with the help of electrical representation of the brain signals is alcoholism. Alcoholism is a serious concern to many in the world as it affects the vital organs of the human body like liver, brain, lungs, heart, blood, immunity levels etc. In the arena of biomedical research, classification of alcoholic subjects from EEG signals is quite a challenging task. In this paper, the alcoholic EEG signals are analyzed comprehensively for a single alcoholic patient and it is classified with many post classifiers. Initially Correlation Dimension features are extracted from the EEG signals and then it is classified with the help of Detrend Fluctuation Analysis (DFA). In order to improve the classification accuracy further, it is again classified with 6 other post classifiers such as Linear Discriminant Analysis (LDA), Kernel LDA, Firefly algorithm, Gaussian Mixture Model (GMM), Logistic Regression (LR) and Softmax Discriminant Classifier (SDC). Results report a high classification accuracy of 97.91% when GMM is employed followed by a classification accuracy of 97.33% when Logistic Regression is employed. A comparatively low classification accuracy of 89.6% is obtained when LDA was employed.
AB - Different pathological and physiological activities of the brain can be analyzed by means of utilizing Electroencephalography (EEG) signals. One such important activity which can be assessed and understood with the help of electrical representation of the brain signals is alcoholism. Alcoholism is a serious concern to many in the world as it affects the vital organs of the human body like liver, brain, lungs, heart, blood, immunity levels etc. In the arena of biomedical research, classification of alcoholic subjects from EEG signals is quite a challenging task. In this paper, the alcoholic EEG signals are analyzed comprehensively for a single alcoholic patient and it is classified with many post classifiers. Initially Correlation Dimension features are extracted from the EEG signals and then it is classified with the help of Detrend Fluctuation Analysis (DFA). In order to improve the classification accuracy further, it is again classified with 6 other post classifiers such as Linear Discriminant Analysis (LDA), Kernel LDA, Firefly algorithm, Gaussian Mixture Model (GMM), Logistic Regression (LR) and Softmax Discriminant Classifier (SDC). Results report a high classification accuracy of 97.91% when GMM is employed followed by a classification accuracy of 97.33% when Logistic Regression is employed. A comparatively low classification accuracy of 89.6% is obtained when LDA was employed.
KW - Alcoholism
KW - DFA
KW - EEG
KW - GMM
KW - LDA
UR - http://www.scopus.com/inward/record.url?scp=85068315033&partnerID=8YFLogxK
U2 - 10.1109/IWW-BCI.2019.8737328
DO - 10.1109/IWW-BCI.2019.8737328
M3 - Conference contribution
AN - SCOPUS:85068315033
T3 - 7th International Winter Conference on Brain-Computer Interface, BCI 2019
BT - 7th International Winter Conference on Brain-Computer Interface, BCI 2019
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 7th International Winter Conference on Brain-Computer Interface, BCI 2019
Y2 - 18 February 2019 through 20 February 2019
ER -