Mental Stress Classification Based on Selected Electroencephalography Channels Using Correlation Coefficient of...

Area of Science:

• Neuroscience
• Biomedical Engineering
• Machine Learning

Background:

• Electroencephalography (EEG) provides insights into brain activity, including responses to mental stress.
• High channel counts in EEG present challenges like computational complexity and setup time.
• Accurate mental stress identification from EEG is crucial for practical applications.

Purpose of the Study:

• To develop a method for identifying and ranking optimal EEG channels for mental stress detection.
• To address limitations of high-density EEG in stress monitoring.
• To enhance the consistency and practicality of EEG-based stress detection.

Main Methods:

• Introduction of the CCHP (Commonly optimal channels for stress detection) method.
• Identification of universally optimal EEG channels based on stress sensitivity.
• Feature extraction from time, frequency, and time-frequency domains.
• Application of machine learning algorithms (RLDA, SVM, KNN).

Main Results:

• Eight EEG channels were identified as universally optimal for detecting stress variances.
• The SVM algorithm achieved a classification accuracy of 81.56% using features from the optimal channels.
• The proposed method demonstrated superior performance compared to existing methodologies.

Conclusions:

• The CCHP method effectively identifies optimal EEG channels for stress detection.
• This approach reduces computational load and setup time, enabling practical applications.
• The findings pave the way for real-time stress detection devices and informed clinical decisions.