A novel EEG-based major depressive disorder detection framework with two-stage feature selection.

Area of Science:

• Neuroscience
• Biomedical Engineering
• Machine Learning

Background:

• Major Depressive Disorder (MDD) significantly impacts daily life and work.
• Accurate and early detection of MDD is crucial for effective treatment.

Purpose of the Study:

• To develop a novel automatic framework for detecting Major Depressive Disorder (MDD) using electroencephalogram (EEG) signals.
• To enhance the accuracy and efficiency of MDD screening and diagnosis.

Main Methods:

• Feature extraction from EEG signals within specific frequency bands.
• A two-stage feature selection method (PAR) combining Pearson correlation coefficient (PCC) and recursive feature elimination (RFE).
• Application of machine learning models including Support Vector Machine (SVM), Logistic Regression (LR), and Linear Regression (LNR) for MDD detection.

Main Results:

• The proposed framework achieved high accuracy (0.9895) and F1-score (0.9846) for MDD detection.
• A high regression determination coefficient (R²) of 0.9479 was obtained for MDD severity assessment.
• Outperformed existing MDD detection methods in terms of accuracy and F1-score.

Conclusions:

• The developed MDD detection framework demonstrates state-of-the-art performance.
• Potential for deployment in medical systems to assist physicians in screening MDD patients.
• Offers a promising tool for objective and efficient MDD diagnosis.