Depressive Disorder Recognition Based on Frontal EEG Signals and Deep Learning

Area of Science:

• Neuroscience
• Artificial Intelligence
• Clinical Psychology

Background:

• Depressive disorder (DD) is a prevalent mental health condition with significant impacts on well-being.
• Current DD diagnosis relies on subjective methods, lacking objective and automated tools.
• High-density electroencephalogram (EEG) use for DD diagnosis is limited by practicality and efficiency concerns.

Purpose of the Study:

• To develop an accurate and practical method for diagnosing depressive disorder (DD) using limited-channel EEG signals.
• To evaluate the efficacy of deep learning models, specifically MRCNN-LSTM and MRCNN-RSE, for DD recognition.
• To explore the relationship between EEG frequency bands and classification performance in DD diagnosis.

Main Methods:

• Collected 10-minute resting-state EEG signals from 41 DD patients and 34 healthy controls (HCs).
• Utilized frontal six-channel EEG data, focusing on specific frequency bands (8-30 Hz).
• Developed and compared two deep learning models: MRCNN-LSTM and MRCNN-RSE for DD classification.

Main Results:

• Higher EEG frequency bands demonstrated superior classification performance for DD diagnosis.
• The MRCNN-RSE model achieved the highest classification accuracy of 98.48 ± 0.22% using 8-30 Hz EEG signals.
• The study highlights the potential of low-density EEG combined with deep learning for practical DD diagnosis.

Conclusions:

• The proposed analytical framework offers an accurate and practical strategy for depressive disorder diagnosis.
• This approach provides essential theoretical and technical support for DD treatment and efficacy evaluation.
• Frontal six-channel EEG combined with deep learning shows promise for objective and efficient DD assessment.