Mental Stress Classification Based on a Support Vector Machine and Naive Bayes Using Electrocardiogram Signals

Area of Science:

• Biomedical Engineering
• Computational Neuroscience
• Psychophysiology

Background:

• Mental health monitoring is essential for preventing conditions like depression.
• Electrocardiogram (ECG) signals contain physiological markers of emotional states.
• Accurate stress level classification can improve mental well-being management.

Purpose of the Study:

• To develop and evaluate a machine learning model for classifying ECG data into four distinct emotional states based on stress levels.
• To enhance stress classification accuracy by incorporating specific ECG interval calculations.

Main Methods:

• Utilized a support vector machine (SVM) with one-against-all and naive Bayes algorithms for classification.
• Calculated average R-S peak, R-R interval, and Q-T interval from ECG data to define stress classification criteria.
• Employed confusion matrix, receiver operating characteristic (ROC) curve, and minimum classification error for performance evaluation.

Main Results:

• The proposed model achieved an average stress classification accuracy of 97.6%.
• The method demonstrated an improvement of 8.7% in accuracy compared to previous stress classification algorithms.
• The selected ECG intervals proved effective in differentiating emotional states.

Conclusions:

• Quantifying stress signals through ECG analysis is feasible and accurate.
• The developed classification model offers a promising tool for objective mental state assessment.
• This approach can contribute to more effective mental health management strategies.