Key result
A hybrid CNN-TCN model applied to raw EEG signals continuously estimated Beck Depression Inventory scores with a mean squared error of 5.64 and mean absolute error of 1.73 in the eyes-open state.
Why the study?
Depression scores are traditionally evaluated qualitatively via questionnaires, motivating the use of raw EEG signals with deep neural networks to continuously estimate Beck depression inventory scores.
Cross-Sectional (n=119)
A hybrid CNN-TCN model can accurately and continuously estimate depression severity scores from raw EEG signals, outperforming existing deep learning and statistical methods.
May enable continuous raw-EEG BDI estimation; leaves open prospective clinical validation before practice adoption.
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.
No takes yet. Share an insight, caveat, or question.
A 2022 study conducted a cross-sectional in Depression (n=119). Hybrid convolutional and temporal-convolutional neural network (CNN-TCN) vs. State-of-the-art deep learning and statistical regression methods was evaluated on Mean squared error (MSE) for continuous estimation of Beck Depression Inventory (BDI) score using raw EEG in the eyes-open state. A hybrid CNN-TCN model applied to raw EEG signals continuously estimated Beck Depression Inventory scores with a mean squared error of 5.64 and mean absolute error of 1.73 in the eyes-open state.
Synapse has enriched 4 closely related papers on similar clinical questions. Consider them for comparative context: