Compact Deep Learning for Major Depressive Disorder Classification from Resting-State EEG: Independent Cohort Validation and Channel-Level Explainability

Abstract

Aims: Major depressive disorder (MDD) is a prevalent psychiatric disorder, and objective neurophysiological measures may complement clinical assessment. This study assessed whether short resting-state EEG segments can distinguish MDD from healthy controls (HC) using two compact deep learning architectures and whether channel ablation provides interpretable spatial information. Materials and Methods: The retrospective dataset included 112 eyes-closed resting-state EEG recordings from 56 individuals with MDD and 56 HC. Ninety-two recordings were used for model development and internal evaluation, while 20 independent recordings (10 MDD, 10 HC) were reserved for external validation. Nineteen channels were standardised to 125 Hz and segmented into non-overlapping 1-s windows. Lightweight 1D Convolutional Neural Network (Light 1D-CNN) and CNN-MiniTransformer models were trained using identical inputs. Channel ablation quantified global and classspecific spatial importance. Results: Internally, Light 1D-CNN achieved 96.43% accuracy, 96.39% F1-score, and 0.9941 AUROC; CNN-MiniTransformer achieved 95.26%, 95.31%, and 0.9906, respectively. On external validation, Light 1D-CNN achieved 93.72% accuracy and 0.9785 AUROC, while CNN-MiniTransformer achieved 94.65% and 0.9872. P4 was the most influential MDD-related channel in both models, followed by Fz and F4, with parietal and frontal regions showing highest importance. Attention-based modelling showed greater external stability while preserving efficient temporal feature extraction. Conclusion: Compact deep learning models maintained strong segment-level discrimination on independent EEG recordings and converged on similar parietal-frontal importance patterns, supporting further participant-level, multicentre validation of interpretable EEG-based MDD classification.

Keywords

• Major depressive disorder;
• deep learning;
• EEG
• explainable artificial intelligence
• ; external validation
• deep learnin

References

1. 1. Duan L, Duan H, Qiao Y, Sha S, Qi S, Zhang X, Huang J, Huang X, Wang C. Machine learning approaches for MDD detection and emotion decoding using EEG signals. Front Hum Neurosci. 2020;14:284. doi:10.3389/fnhum.2020.00284.
2. 2. Wang Y, Zhao S, Jiang H, Li S, Li T, Pan G. M-MDD: A multi-task deep learning framework for major depressive disorder diagnosis using EEG. Neurocomputing. 2025;636:130008. doi:10.1016/j.neucom.2025.130008.
3. 3. Xia M, Zhang Y, Wu Y, Wang X. An end-to-end deep learning model for EEG-based major depressive disorder classification. IEEE Access. 2023;11:41337-41347. doi:10.1109/ACCESS.2023.3270426.
4. 4. Anik IA, Kamal AHM, Kabir MA, Uddin S, Moni MA. A robust deep-learning model to detect major depressive disorder utilizing EEG signals. IEEE Trans Artif Intell. 2024;5(10):4938-4947. doi:10.1109/TAI.2024.3394792.
5. 5. Biasiucci A, Franceschiello B, Murray MM. Electroencephalography. Curr Biol. 2019;29(3):R80-R85. doi:10.1016/j.cub.2018.11.052.
6. 6. Jaseja H. Electroencephalography in the diagnosis and management of treatment-resistant depression with comorbid epilepsy: A novel strategy. Gen Psychiatry. 2023;36(2):e100868. doi:10.1136/gpsych-2022-100868.
7. 7. Wang Y, et al. DiffMDD: A diffusion-based deep learning framework for MDD diagnosis using EEG. IEEE Trans Neural Syst Rehabil Eng. 2024;32:728-738. doi:10.1109/TNSRE.2024.3360465.
8. 8. Safayari A, Bolhasani H. Depression diagnosis by deep learning using EEG signals: A systematic review. Med Nov Technol Devices. 2021;12:100102. doi:10.1016/j.medntd.2021.100102.

9. 9. Rehman H, Khan DM, Amanullah H, Kamran L, Rehman OU, Siddiqui ST, Masroor K. Advances in EEG-based detection of major depressive disorder using shallow and deep learning techniques: A systematic review. Comput Biol Med. 2025;192:110154. doi:10.1016/j.compbiomed.2025.110154.
10. 10. Saeedi A, Saeedi M, Maghsoudi A, Shalbaf A. Major depressive disorder diagnosis based on effective connectivity in EEG signals: A convolutional neural network and long short-term memory approach. Cogn Neurodyn. 2021;15(2):239-252. doi:10.1007/s11571-020-09619-0.
11. 11. Loh HW, Ooi CP, Aydemir E, Tuncer T, Dogan S, Acharya UR. Decision support system for major depression detection using spectrogram and convolutional neural network with EEG signals. Expert Syst. 2022;39(3):e12773.
12. 12. Ksibi A, Zakariah M, Menzli LJ, Saidani O, Almuqren L, Hanafieh RAM. Electroencephalography-based depression detection using multiple machine learning techniques. Diagnostics. 2023;13(10):1779.
13. 13. Yu D, Li P, Su F, Wang K, Duan S, Yuan D. Automatic detection of depression using a CNN-Transformer model based on low-channel EEG data. Turk J Electr Eng Comput Sci. 2025;33(5):594-612. doi:10.55730/1300-0632.4146.
14. 14. Liu X, Zhou L, Zhang W, Shao Z, Liu X, Chen C. LightDepressionNet: A lightweight depression diagnostic system based on CNN-Transformer. In: 2025 15th International Conference on Biomedical Engineering and Technology (ICBET); 2025. p. 6-10. IEEE. doi:10.1109/ICBET66094.2025.11470277.
15. 15. Gan W, Zhao R, Ma Y, Ning X. TSF-MDD: A deep learning approach for electroencephalography-based diagnosis of major depressive disorder with temporal-spatial-frequency feature fusion. Bioengineering. 2025;12(2):95. doi:10.3390/bioengineering12020095.
16. 16. Balakrishnan S, Murugesan RK, Ashraf H, Lim EL, Faisal A. EEGMetaNet: An enhanced Major Depressive Disorder detection framework using electroencephalogram and one-dimensional convolutional neural network. Appl Sci. 2026;16(15):7564. doi:10.3390/app16157564.
17. 17. Naeim M, Atadokht A. Explainable temporal deep learning for EEG-based depression detection using resting-state brain dynamics. Int J Methods Psychiatr Res. 2026;35(2):e70088.
18. 18. Liu S, Cui Y, Xu Y, Zhou B, Li G, Luo X. From EEG signals to quantitative assessment: Predicting depression severity using a novel deep learning framework. Sci Rep. 2026.
19. 19. Rahmati M, Jalaeianbanayan A, Vahedi J, et al. Novel EEG-based diagnostic framework for Major Depressive Disorder using microstate and entropy features. Cogn Neurodyn. 2025;19:116. doi:10.1007/s11571-025-10305-2.
20. 20. Zheng L, Fu J, Sun Z, Han Y, Shao T, Wang Q, Lin Z. Multi-scale EEG analysis identifies neural circuit signatures of iTBS responsiveness in major depressive disorder. NeuroImage. 2026:121697.
21. 21. Peng D, Zheng WL, Lu BL. Enhancing depression detection from emotion EEG with temporal-spatial-spectral representation learning. In: 2025 47th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC); 2025. p. 1-5. IEEE. doi:10.1109/EMBC58623.2025.11253710.
22. 22. Kozulin I, Merkulova E, Savostyanov V, Shi H, Wang X, Bocharov A, Savostyanov A. Predicting major depressive disorder using neural networks from spectral measures of EEG data. Bioengineering. 2025;12(11):1251.