Alphonsa Sini P. J.; Sherly K. K.

An Optimized Graph-based Deep Learning Framework for Depression Detection Using EEG Signals

PDF (1344KB), PP.85-104

Views: 0 Downloads: 0

Author(s)

Alphonsa Sini P. J. ^1,* Sherly K. K. ²

1. APJ Abdul Kalam Technological University, Rajagiri School of Engineering & Technology, Kerala, India

2. Department of Artificial Intelligence, Rajagiri School of Engineering & Technology, Kerala, India

* Corresponding author.

DOI: https://doi.org/10.5815/ijitcs.2026.04.06

Received: 16 Feb. 2026 / Revised: 10 Apr. 2026 / Accepted: 27 May 2026 / Published: 8 Aug. 2026

Index Terms

EEG, Depression Detection, Graph Convolutional Network, Modified Addax Optimization Algorithm, Efficientnetv2, Spiking Neural Network, Deep Learning

Abstract

Depression is a serious psychiatric disorder that greatly impacts the quality of life and daily functioning of a person. Accurate diagnosis at the early stage is critical for success with intervention. Electroencephalography (EEG) offers a non-invasive technique to assess neurophysiological activity and is thus an important instrument for diagnosis of depression. Current EEG-based deep learning approaches are beset by high-dimensional data, poor feature selection, and poor classification performance owing to the nature of the EEG signal. To address these issues, we introduce EEGEffV2-SpikeNet a new framework for depression detection that combines statistical feature extraction with deep feature extraction through a Graph Convolutional Network (GCN) approach. The proposed model incorporates a new fusion of statistical feature extraction and GCN-based deep feature learning for the extraction of both spatial and temporal EEG features. The extracted features are then optimized by Modified Addax Optimization Algorithm (MAOA), which is a cutting-edge bio-inspired optimization algorithm for optimizing feature selection efficiency by discarding redundant information and improving classification accuracy. For depression classification, EfficientNetV2, Deep Belief Network (DBN) and a Spiking Neural Network (SNN) are utilized based on the computational efficiency of EfficientNetV2 and the biologically simulated processing of SNN for enhancing feature representation and decision-making. Experimental results on two standard EEG datasets validate the better performance of the model, achieving 98.74% accuracy on Dataset 1 and 97.88% accuracy on Dataset 2, outperforming baseline models like DBN, EfficientNet, and SNN. The results prove the framework's promise as a dependable tool for objective and early depression diagnosis, with clinical application and mental health monitoring implications.

Cite This Paper

Alphonsa Sini P. J., Sherly K. K., "An Optimized Graph-based Deep Learning Framework for Depression Detection Using EEG Signals", International Journal of Information Technology and Computer Science(IJITCS), Vol.18, No.4, pp.85-104, 2026. DOI:10.5815/ijitcs.2026.04.06

Reference

[1]J. Fang, G. Li, W. Xu, W. Liu, G. Chen, Y. Zhu, Y. Luo, X. Luo, and B. Zhou, “Exploring abnormal brain functional connectivity in healthy adults, depressive disorder, and generalized anxiety disorder through EEG signals: A machine learning approach for triple classification,” Brain Sciences, vol. 14, no. 3, p. 245, 2024.
[2]A. A. Chandanwala, S. Bhowmik, P. Chaudhury, and S. C. Pravin, “Hybrid quantum deep learning model for emotion detection using raw EEG signal analysis,” arXiv preprint arXiv:2411.17715, 2024.
[3]A. Megala, R. D. Patil, V. G. Pushparathi, G. Manoharan, S. Hemambika, and B. Khan, “Experimental analysis of employee depression identification system using modified learning procedure with AI assistance,” in Proc. 5th Int. Conf. Intelligent Communication Technologies and Virtual Mobile Networks (ICICV), pp. 282–289, 2024.
[4]X. Chen, Y. Kong, H. Chang, Y. Gao, Z. Liu, J. L. Coatrieux, and H. Shu, “MGSN: Depression EEG lightweight detection based on multiscale DGCN and SNN for multichannel topology,” Biomedical Signal Processing and Control, vol. 92, p. 106051, 2024.
[5]F. Zhang, C. Yang, L. You, X. Wang, Y. Yuan, R. L. B. Jeannès, H. Shu, and W. Xiang, “WS-BiLSTM-MA: Wavelet scattering-based BiLSTM with mixed attention block for MDD recognition using multi-channel EEG signals,” IEEE Trans. Instrum. Meas., 2024.
[6]X. Jia, J. Chen, K. Liu, Q. Wang, and J. He, “Multimodal depression detection based on an attention graph convolution and transformer,” Math. Biosci. Eng., vol. 22, no. 3, pp. 652–676, 2025.
[7]S. Khan, S. M. Umar Saeed, J. Frnda, A. Arsalan, R. Amin, R. Gantassi, and S. H. Noorani, “A machine learning based depression screening framework using temporal domain features of electroencephalography signals,” PLOS ONE, vol. 19, no. 3, p. e0299127, 2024.
[8]A. Ksibi, M. Zakariah, L. J. Menzli, O. Saidani, L. Almuqren, and R. A. M. Hanafieh, “Electroencephalography-based depression detection using multiple machine learning techniques,” Diagnostics, vol. 13, no. 10, p. 1779, 2023.
[9]Z. Ruan, P. Yang, J. Huang, K. Yang, Y. Lv, and Z. Q. Zhang, “Automatic depression detection among higher education students based on DeepFM,” IEEE Trans. Instrum. Meas., 2024.
[10]Z. Fan, F. Chen, X. Xia, and Y. Liu, “EEG emotion classification based on graph convolutional network,” Appl. Sci., vol. 14, no. 2, p. 726, 2024.
[11]M. J. Nazari, M. Shalbafan, N. Eissazade, E. Khalilian, Z. Vahabi, N. Masjedi, S. S. Ghidary, M. Saadat, and S. A. Sadegh-Zadeh, “A machine learning approach for differentiating bipolar disorder type II and borderline personality disorder using electroencephalography,” PLOS ONE, vol. 19, no. 6, p. e0303699, 2024.
[12]K. Stefanou, K. D. Tzimourta, C. Bellos, G. Stergios, K. Markoglou, E. Gionanidis, M. G. Tsipouras, N. Giannakeas, A. T. Tzallas, and A. Miltiadous, “A novel CNN-based framework for Alzheimer’s disease detection using EEG spectrogram representations,” J. Pers. Med., vol. 15, no. 1, p. 27, 2025.
[13]S. Khan, S. M. U. Saeed, M. Rehman, S. H. Noorani, and U. Rauf, Temporal domain feature-based machine learning framework for depression screening using EEG signals.
[14]M. Pettorruso, G. Di Lorenzo, B. Benatti, G. D’Andrea, C. Cavallotto, R. Carullo, G. Mancusi, O. Di Marco, G. Mammarella, A. D’Attilio, and E. Barlocci, “Overcoming treatment-resistant depression with machine-learning based tools,” Front. Psychiatry, vol. 15, p. 1436006, 2024.
[15]S. Bhadra, C. J. Kumar, and D. K. Bhattacharyya, “Multiview EEG signal analysis for diagnosis of schizophrenia: An optimized deep learning approach,” Multimed. Tools Appl., 2024.
[16]X. Li, Y. Dong, Y. Yi, Z. Liang, and S. Yan, “Hypergraph neural network for multimodal depression recognition,” Electronics, vol. 13, no. 22, p. 4544, 2024.
[17]M. Karthiga, E. Suganya, S. Sountharrajan, B. Balusamy, and S. Selvarajan, “EEG-based smart emotion recognition using meta-heuristic optimization and hybrid deep learning techniques,” Sci. Rep., vol. 14, 2024.
[18]X. Liu, S. Qu, G. Luo, C. Yan, D. Wang, N. Chu, F. Tian, J. Zhu, X. Li, S. Sun, and B. Hu, “A generalized depression recognition framework based on cross-center and cross-task EEG signals,” Biomed. Signal Process. Control, vol. 104, p. 107492, 2025.
[19]A. M. H. Kwok, J. Cheong, S. Kalkan, and H. Gunes, “Machine learning fairness for depression detection using EEG data,” arXiv preprint arXiv:2501.18192, 2025.
[20]A. Tavanaei and A. S. Maida, “Bio-inspired spiking convolutional neural network using layer-wise sparse coding and STDP learning,” arXiv preprint arXiv:1611.03000, 2016.
[21]F. Alenizi, B. Bouallegue, A. Sam, and R. Boostani, “EEG-based depression classification and brain region analysis using a hybrid NeuCube and dictionary learning framework,” Authorea Preprints, 2024.
[22]S. Liu, L. An, and Z. Jia, “A Domain Adversarial Learning Framework for Major Depression Disorder Diagnosis,” In ICASSP. IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) IEEE, (pp. 1-5) 2025-2025.
[23]K. H. Chung, Y. S. Chang, W. T. Yen, L. Lin, and S. Abimannan, “Depression assessment using integrated multi-featured EEG bands deep neural network models,” Comput. Struct. Biotechnol. J., vol. 23, pp. 1450–1468, 2024.
[24]W. Gan, R. Zhao, Y. Ma, and X. Ning, “TSF-MDD: A deep learning approach for EEG-based diagnosis with temporal–spatial–frequency feature fusion,” Bioengineering, vol. 12, no. 2, p. 95, 2025.
[25]W. Chen, Y. Liao, R. Dai, Y. Dong, and L. Huang, “EEG-based emotion recognition using graph convolutional neural network with dual attention mechanism,” Front. Comput. Neurosci., vol. 18, 2024.
[26]J. Li, G. Feng, J. Lv, Y. Chen, R. Chen, F. Chen, S. Zhang, M. I. Vai, S. H. Pun, and P. U. Mak, “A lightweight multi-mental disorders detection method using entropy-based matrix from single-channel EEG signals,” Brain Sciences, vol. 14, no. 10, p. 987, 2024.
[27]M. H. Abidi, K. Moiduddin, R. Ayub, M. K. Mohammed, A. Shankar, and S. Shiaeles, “EEGDepressionNet: A novel self-attention-based gated DenseNet model for depression detection,” IEEE J. Biomed. Health Inform., 2024.
[28]W. Liu, K. Jia, and Z. Wang, “Graph-based EEG approach for depression prediction,” Front. Neurosci., vol. 18, 2024.
[29]J. Shen, K. Li, H. Liang, Z. Zhao, Y. Ma, J. Wu, J. Zhang, Y. Zhang, and B. Hu, “HEMAsNet: A hemisphere asymmetry network for depression recognition from EEG signals,” IEEE J. Biomed. Health Inform., 2024.
[30]T. Xing, Y. Dou, X. Chen, J. Zhou, X. Xie, and S. Peng, “An adaptive multi-graph neural network with multimodal feature fusion for MDD detection,” Sci. Rep., vol. 14, 2024.
[31]P. Penava and R. Buettner, “Early-stage non-severe depression detection using a convolutional neural network based on EEG data,” IEEE Access, 2024.
[32]N. Bhanja, S. K. Dhara, and P. Khampariya, “An efficient deep learning with an optimization framework to analyze EEG signals,” Multimed. Tools Appl., 2024.
[33]M. Ying, J. Zhu, X. Li, and B. Hu, “A functional connectivity-based model with a lightweight attention mechanism for depression recognition,” IEEE Trans. Neural Syst. Rehabil. Eng., 2024.
[34]S. De, A. Singh, V. Tiwari, H. Patel, G. N. Vivekananda, and D. S. Rajput, “SLiTRANet: An EEG-based automated diagnosis framework using LGCN and transformer,” IEEE Access, 2024.
[35]T. Hamadneh, K. Kaabneh, O. Alssayed, K. Eguchi, S. Gochhait, I. Leonova, and M. Dehghani, “Addax optimization algorithm: A novel nature-inspired optimizer for engineering applications,” Int. J. Intell. Eng. Syst., vol. 17, no. 3, 2024.

International Journal of Information Technology and Computer Science (IJITCS)