EEG Classification of Normal and Alcoholic by Deep Learning.

Area of Science:

• Neuroscience
• Medical Diagnostics
• Artificial Intelligence

Background:

• Alcohol dependence is a prevalent global mental health issue with severe health consequences.
• Current diagnostic methods for alcoholism lack standardized testing and often rely on manual analysis of electroencephalogram (EEG) signals.
• Deep learning offers automated feature extraction from EEG, surpassing traditional machine learning's reliance on human intervention.

Purpose of the Study:

• To develop and evaluate a novel deep learning model for the automatic classification and diagnosis of alcoholism using EEG signals.
• To address the limited research in applying deep learning to EEG-based alcoholism detection.

Main Methods:

• A multilayer discrete wavelet transform was employed for denoising EEG data.
• A convolutional neural network (CNN) combined with a bidirectional long short-term memory (BiLSTM) network was utilized for automated feature extraction.
• The processed EEG signals were classified to diagnose alcohol dependence.

Main Results:

• The proposed deep learning method demonstrated effective automatic feature extraction and classification of alcohol EEG signals.
• The diagnostic accuracy achieved was 99.32%, significantly outperforming existing algorithms.
• The method provides a more convenient and potentially more accurate approach to diagnosing alcoholism.

Conclusions:

• The developed deep learning model offers a highly accurate and automated solution for diagnosing alcoholism via EEG analysis.
• This approach represents a significant advancement over traditional methods, paving the way for improved clinical diagnostic tools.