Continuous Scoring of Depression From EEG Signals via a Hybrid of Convolutional Neural Networks

• Published in: IEEE transactions on neural systems and rehabilitation engineering Vol. 30; pp. 176 - 183
• Main Authors: Hashempour, S., Boostani, R., Mohammadi, M., Sanei, S.
• Format: Journal Article
• Language: English
• Published: United States IEEE 2022; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Artificial neural networks; Beck depression test; CNN; Convolution; Convolutional neural networks; Deep learning; Depression; EEG; Electrodes; Electroencephalography; Electroencephalography - methods; Feature extraction; Humans; Neural networks; Neural Networks, Computer; Qualitative analysis; Scalp; Statistical analysis; Statistical methods; Statistics; TCN
• ISSN: 1534-4320; 1558-0210; 1558-0210
• DOI: 10.1109/TNSRE.2022.3143162

Depression score is traditionally determined by taking the Beck depression inventory (BDI) test, which is a qualitative questionnaire. Quantitative scoring of depression has also been achieved by analyzing and classifying pre-recorded electroencephalography (EEG) signals. Here, we go one step further and apply raw EEG signals to a proposed hybrid convolutional and temporal-convolutional neural network (CNN-TCN) to continuously estimate the BDI score. In this research, the EEG signals of 119 individuals are captured by 64 scalp electrodes through successive eyes-closed and eyes-open intervals. Moreover, all the subjects take the BDI test and their scores are determined. The proposed CNN-TCN provides mean squared error (MSE) of 5.64±1.6 and mean absolute error (MAE) of 1.73±0.27 for eyes-open state and also provides MSE of 9.53±2.94 and MAE of 2.32±0.35 for the eyes-closed state, which significantly surpasses state-of-the-art deep network methods. In another approach, conventional EEG features are elicited from the EEG signals in successive frames and apply them to the proposed CNN-TCN in conjunction with known statistical regression methods. Our method provides MSE of 10.81±5.14 and MAE of 2.41±0.59 that statistically outperform the statistical regression methods. Moreover, the results with raw EEG are significantly better than those with EEG features.