EEG-Based Depression Detection Using Welch Power Spectral Density Analysis and Transformer Learning

Authors

DOI:

https://doi.org/10.26877/asset.v8i4.3740

Keywords:

Electroencephalography (EEG), Depression Detection, Welch Method, Power Spectral Density, Convolutional Neural Networks, Recurrent Neural Networks, Transformer Learning

Abstract

Depression diagnosis remains largely dependent on subjective clinical assessments, creating a strong need for objective and reproducible biomarkers. Electroencephalography (EEG) offers a non-invasive and cost-effective solution for capturing neural activity associated with depressive states. This study proposes a deep learning-based framework for automated depression detection using EEG Power Spectral Density (PSD) features estimated through the Welch method. After preprocessing with bandpass filtering, notch filtering, and Independent Component Analysis (ICA)-based artifact removal, PSD features were extracted from the delta, theta, alpha, beta, and gamma bands. These spectral representations were then classified using Convolutional Neural Networks (CNN), Recurrent Neural Networks (RNN), and Transformer-based architectures. Experimental results on EEG recordings from 90 subjects (60 depressed and 30 healthy controls) demonstrated that the Transformer model achieved a classification accuracy of 83.78%, with favorable convergence, generalization, and computational efficiency compared with the CNN and RNN models. The findings confirm that Welch-based spectral analysis combined with Transformer learning provides an effective and scalable framework for objective depression detection, with strong potential for clinical decision support and portable mental health monitoring systems.

Author Biographies

  • Pratima Singh, Galgotias University

    School of Computer Applications and Technology, Galgotias University, Greater Noida 203201, India

  • Alok Singh Chauhan, Galgotias University

    School of Computer Applications and Technology, Galgotias University, Greater Noida 203201, India

    INTI International University, 71800 Nilai, Negeri Sembilan, Malaysia

  • Dhyanendra Jain, ABES Engineering College

    Department of CSE-AIML, ABES Engineering College, Ghaziabad, India

  • Ashok Kumar Rai, Bennett University

    School of Computer Science Engineering and Technology, Bennett University, Greater Noida, India

  • Palak Gupta, Galgotias University

    School of Business, Galgotias University, Greater Noida 203201, India

  • Rajiv Chourasiya, Galgotias University

    School of Computer Applications and Technology, Galgotias University, Greater Noida 203201, India

References

[1] Čukić M, López V, Pavón J. Classification of Depression Through Resting-State Electroencephalogram as a Novel Practice in Psychiatry: Review. J Med Internet Res 2020;22:e19548. https://doi.org/10.2196/19548.

[2] Kennis M, Gerritsen L, Van Dalen M, et al. Prospective biomarkers of major depressive disorder: a systematic review and meta-analysis. Mol Psychiatry 2020;25:321–38. https://doi.org/10.1038/s41380-019-0585-z.

[3] Thibodeau R, Jorgensen RS, Kim S. Depression, anxiety, and resting frontal EEG asymmetry: A meta-analytic review. Journal of Abnormal Psychology 2006;115:715–29. https://doi.org/10.1037/0021-843X.115.4.715.

[4] Nosek BA, Hawkins CB, Frazier RS. Implicit social cognition: from measures to mechanisms. Trends in Cognitive Sciences 2011;15:152–9. https://doi.org/10.1016/j.tics.2011.01.005.

[5] Wu C-T, Dillon DG, Hsu H-C, et al. Depression Detection Using Relative EEG Power Induced by Emotionally Positive Images and a Conformal Kernel Support Vector Machine. Applied Sciences 2018;8:1244. https://doi.org/10.3390/app8081244.

[6] Acharya UR, Oh SL, Hagiwara Y, et al. Automated EEG-based screening of depression using deep convolutional neural network. Computer Methods and Programs in Biomedicine 2018;161:103–13. https://doi.org/10.1016/j.cmpb.2018.04.012.

[7] Liu W, Jia K, Wang Z, et al. A Depression Prediction Algorithm Based on Spatiotemporal Feature of EEG Signal. Brain Sciences 2022;12:630. https://doi.org/10.3390/brainsci12050630.

[8] Welch P. The use of fast Fourier transform for the estimation of power spectra: A method based on time averaging over short, modified periodograms. IEEE Trans Audio Electroacoust 1967;15:70–3. https://doi.org/10.1109/TAU.1967.1161901.

[9] Pfister H. Discrete-time signal processing. Lecture Note, Pfister Ee Duke Edu/Courses/Ece485/Dtsp Pdf 2017;4:2. https://pfister.ee.duke.edu/courses/ece485/dtsp.pdf. Accessed 5 April 2026.

[10] Stoica P, Moses RL. Spectral analysis of signals. vol. 452. Pearson Prentice Hall Upper Saddle River, NJ; 2005. https://doi.org/10.1016/B978-013113956-5.X5000-7.

[11] Wang Y, Liu F, Yang L. EEG-Based Depression Recognition Using Intrinsic Time-scale Decomposition and Temporal Convolution Network. The Fifth International Conference on Biological Information and Biomedical Engineering, Hangzhou China: ACM; 2021, p. 1–6. https://doi.org/10.1145/3469678.3469683.

[12] Liu W, Jia K, Wang Z. Graph-based EEG approach for depression prediction: integrating time-frequency complexity and spatial topology. Front Neurosci 2024;18:1367212. https://doi.org/10.3389/fnins.2024.1367212.

[13] Wang B, Kang Y, Huo D, et al. EEG diagnosis of depression based on multi-channel data fusion and clipping augmentation and convolutional neural network. Front Physiol 2022;13:1029298. https://doi.org/10.3389/fphys.2022.1029298.

[14] Liu B, Chang H, Peng K, et al. An End-to-End Depression Recognition Method Based on EEGNet. Front Psychiatry 2022;13:864393. https://doi.org/10.3389/fpsyt.2022.864393.

[15] Li X, La R, Wang Y, et al. A Deep Learning Approach for Mild Depression Recognition Based on Functional Connectivity Using Electroencephalography. Front Neurosci 2020;14:192. https://doi.org/10.3389/fnins.2020.00192.

[16] Umair M, Ahmad J, Alasbali N, et al. Decentralized EEG-based detection of major depressive disorder via transformer architectures and split learning. Front Comput Neurosci 2025;19:1569828. https://doi.org/10.3389/fncom.2025.1569828.

[17] Mouazen B, Benali A, Chebchoub NT, et al. Enhancing EEG-Based Emotion Detection with Hybrid Models: Insights from DEAP Dataset Applications. Sensors 2025;25:1827. https://doi.org/10.3390/s25061827.

[18] Tan Q, Miao M. Detection of Depression in EEG Signals Based on Convolutional Transformer and Adaptive Transfer Learning. In: Shi Z, Witbrock M, Tian Q, editors. Intelligence Science V, vol. 720, Cham: Springer Nature Switzerland; 2025, p. 30–9. https://doi.org/10.1007/978-3-031-71253-1_3.

[19] Lawhern VJ, Solon AJ, Waytowich NR, et al. EEGNet: a compact convolutional neural network for EEG-based brain–computer interfaces. Journal of Neural Engineering 2018;15:056013. https://doi.org/10.1088/1741-2552/aace8c.

[20] Schirrmeister RT, Springenberg JT, Fiederer LDJ, et al. Deep learning with convolutional neural networks for EEG decoding and visualization. Human Brain Mapping 2017;38:5391–420. https://doi.org/10.1002/hbm.23730.

[21] Song X, Yan D, Zhao L, et al. LSDD-EEGNet: An efficient end-to-end framework for EEG-based depression detection. Biomedical Signal Processing and Control 2022;75:103612. https://doi.org/10.1016/j.bspc.2022.103612.

[22] Qayyum A, Razzak I, Tanveer M, et al. High-Density Electroencephalography and Speech Signal Based Deep Framework for Clinical Depression Diagnosis. IEEE/ACM Trans Comput Biol and Bioinf 2023;20:2587–97. https://doi.org/10.1109/TCBB.2023.3257175.

[23] Tasci G, Loh HW, Barua PD, et al. Automated accurate detection of depression using twin Pascal’s triangles lattice pattern with EEG Signals. Knowledge-Based Systems 2023;260:110190. https://doi.org/10.1016/j.knosys.2022.110190.

[24] Dehghani M, Asayesh V, Nikjeh MT, et al. Investigation Capability of EEG-Based Non-Linear Features in Depression Detection. Fbt 2025. https://doi.org/10.18502/fbt.v12i4.19809.

[25] Liu X, Qu S, Luo G, et al. A generalized depression recognition framework based on cross-center and cross-task EEG signals. Biomedical Signal Processing and Control 2025;104:107492. https://doi.org/10.1016/j.bspc.2025.107492.

[26] Lau ZJ, Pham T, Chen SHA, et al. Brain entropy, fractal dimensions and predictability: A review of complexity measures for EEG in healthy and neuropsychiatric populations. Eur J of Neuroscience 2022;56:5047–69. https://doi.org/10.1111/ejn.15800.

[27] Cukic M, Pokrajac D, Stokic M, et al. EEG machine learning with Higuchi fractal dimension and Sample Entropy as features for successful detection of depression 2018. https://doi.org/10.48550/ARXIV.1803.05985.

[28] Avots E, Jermakovs K, Bachmann M, et al. Ensemble approach for detection of depression using EEG features 2021. https://doi.org/10.48550/ARXIV.2103.08467.

[29] Radenković MČ, Lopez VL. Machine Learning Approaches for Detecting the Depression from Resting-State Electroencephalogram (EEG): A Review Study 2019. https://doi.org/10.48550/ARXIV.1909.03115.

[30] Thoduparambil PP, Dominic A, Varghese SM. EEG-based deep learning model for the automatic detection of clinical depression. Phys Eng Sci Med 2020;43:1349–60. https://doi.org/10.1007/s13246-020-00938-4.

[31] Shen A, Sun J, Chen X, et al. A data-centric and interpretable EEG framework for depression severity grading using SHAP-based insights. J NeuroEngineering Rehabil 2025;22:116. https://doi.org/10.1186/s12984-025-01645-5.

[32] Kwok AMH, Cheong J, Kalkan S, et al. Machine Learning Fairness for Depression Detection Using EEG Data. 2025 IEEE 22nd International Symposium on Biomedical Imaging (ISBI), Houston, TX, USA: IEEE; 2025, p. 1–5. https://doi.org/10.1109/ISBI60581.2025.10980986.

[33] Bagherzadeh S, Norouzi MR, Ghasri A, et al. Automated depression detection via cloud based EEG analysis with transfer learning and synchrosqueezed wavelet transform. Sci Rep 2025;15:18008. https://doi.org/10.1038/s41598-025-02452-7.

[34] Krisna JIT, Luthfiarta A, Cahya LD, et al. Comparing Optimizer Strategies For Enhancing Emotion Classification In IndoBERT Models. ASSET 2024;6:0240203. https://doi.org/10.26877/asset.v6i2.18228.

Downloads

Published

2026-10-12

Issue

Vol. 8 No. 4 (2026): August-October

Section

Articles

License

Copyright (c) 2026 Advance Sustainable Science Engineering and Technology

Creative Commons License

This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.