A novel Alcoholic EEG signals Classification Approach Based on AdaBoost k-means Coupled with Statistical Model

Diykh, Mohammed and Abdulla, Shahab ORCID: https://orcid.org/0000-0002-1193-6969 and Oudah, Atheer Y. and Marhoon, Haydar Abdulameer and Siuly, Siuly ORCID: https://orcid.org/0000-0003-2491-0546 (2021) A novel Alcoholic EEG signals Classification Approach Based on AdaBoost k-means Coupled with Statistical Model. In: 10th International Conference on Health Information Science (HIS 2021), 25 Oct - 28 Oct 2021, Melbourne, Australia.


Abstract

Identification of alcoholism is an important task because it affects the operation of the brain. Alcohol consumption, particularly heavier drinking is identified as an essential factor to develop health issues, such as high blood pressure, immune disorders, and heart diseases. To support health professionals in diagnosis disorders related with alcoholism with a high rate of accuracy, there is an urgent demand to develop an automated expert systems for identification of alcoholism. In this study, an expert system is proposed to identify alcoholism from multi-channel EEG signals. EEG signals are partitioned into small epochs, with each epoch is further divided into sub-segments. A covariance matrix method with its eigenvalues is utilised to extract representative features from each sub-segment. To select most relevant features, a statistic approach named Kolmogorov–Smirnov test is adopted to select the final features set. Finally, in the classification part, a robust algorithm called AdaBoost k-means (AB-k-means) is designed to classify EEG features into two categories alcoholic and non-alcoholic EEG segments. The results in this study show that the proposed model is more efficient than the previous models, and it yielded a high classification rate of 99%. In comparison with well-known classification algorithms such as K-nearest k-means and SVM on the same databases, our proposed model showed a promising result compared with the others. Our findings showed that the proposed model has a potential to implement in automated alcoholism detection systems to be used by experts to provide an accurate and reliable decisions related to alcoholism.


Statistics for USQ ePrint 44781
Statistics for this ePrint Item
Item Type: Conference or Workshop Item (Commonwealth Reporting Category E) (Paper)
Refereed: Yes
Item Status: Live Archive
Faculty/School / Institute/Centre: Current - USQ College (8 Jun 2020 -)
Faculty/School / Institute/Centre: Current - USQ College (8 Jun 2020 -)
Date Deposited: 13 Feb 2022 23:11
Last Modified: 23 Nov 2022 23:59
Uncontrolled Keywords: alcoholism, EEG, AdaBoost k-means, covariance matrix, Kolmogorov–Smirnov
Fields of Research (2008): 11 Medical and Health Sciences > 1199 Other Medical and Health Sciences > 119999 Medical and Health Sciences not elsewhere classified
Fields of Research (2020): 42 HEALTH SCIENCES > 4299 Other health sciences > 429999 Other health sciences not elsewhere classified
Socio-Economic Objectives (2008): C Society > 92 Health > 9299 Other Health > 929999 Health not elsewhere classified
E Expanding Knowledge > 97 Expanding Knowledge > 970111 Expanding Knowledge in the Medical and Health Sciences
Socio-Economic Objectives (2020): 20 HEALTH > 2099 Other health > 209999 Other health not elsewhere classified
Identification Number or DOI: https://doi.org/10.1007/978-3-030-90885-0_8
URI: http://eprints.usq.edu.au/id/eprint/44781

Actions (login required)

View Item Archive Repository Staff Only