{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,19]],"date-time":"2026-10-19T20:40:19Z","timestamp":1792442419709,"version":"4.5.0"},"reference-count":61,"publisher":"JMIR Publications Inc.","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["JMIR AI"],"abstract":"Abstract<\/jats:title>\n \n Background<\/jats:title>\n Depression is a common mental disorder that seriously affects mental health and quality of life. Traditional depression diagnosis mainly relies on clinical interviews and subjective assessments, which may lead to variability in diagnostic outcomes. Electroencephalogram (EEG)-based computer-aided diagnosis has emerged as a promising, objective approach for depression recognition. However, effectively modeling the complex nonlinear characteristics of EEG signals remains a major challenge for existing deep learning methods.<\/jats:p>\n <\/jats:sec>\n \n Objective<\/jats:title>\n This study aimed to investigate the applicability of Kolmogorov-Arnold Networks (KANs) for EEG-based depression recognition and to evaluate the effectiveness of KAN-based nonlinear modeling frameworks on frequency-domain EEG features.<\/jats:p>\n <\/jats:sec>\n \n Methods<\/jats:title>\n \n KAN was introduced into EEG-based depression recognition to enhance nonlinear feature representation of brain electrical signals. Two representative KAN-based convolutional architectures, ConvKAN and its efficiency-optimized variant ConvKAN Efficient, were systematically investigated for EEG classification tasks. These architectures incorporated KANLinear layers together with dynamic grid updating, enhanced weight initialization, and regularization strategies to improve nonlinear modeling capability while maintaining computational efficiency. Extensive experiments were conducted on the EDRA dataset, and the proposed framework was compared with conventional convolutional neural network\u2013based methods and traditional machine learning approaches. Model performance was evaluated using accuracy, precision, recall, and\n F<\/jats:italic>\n 1<\/jats:sub>\n -score metrics.\n <\/jats:p>\n <\/jats:sec>\n \n Results<\/jats:title>\n \n Experimental results demonstrated that the KAN-based frameworks achieved superior EEG classification performance compared with conventional convolutional neural network\u2013based methods and traditional machine learning approaches. Among the evaluated models, ConvKAN Efficient achieved the best overall performance, with accuracy, precision, recall, and\n F<\/jats:italic>\n 1<\/jats:sub>\n -score all exceeding 94% on the EDRA dataset. The results indicate that KAN-based architectures can effectively capture the complex nonlinear characteristics of EEG signals while preserving high computational efficiency.\n <\/jats:p>\n <\/jats:sec>\n \n Conclusions<\/jats:title>\n This study systematically demonstrates the applicability of KAN-based nonlinear modeling frameworks to EEG-based depression recognition. The proposed framework provides a scalable and effective approach for EEG signal analysis and offers valuable technical support for both theoretical research and clinical applications in computer-aided depression recognition.<\/jats:p>\n <\/jats:sec>","DOI":"10.2196\/101076","type":"journal-article","created":{"date-parts":[[2026,10,19]],"date-time":"2026-10-19T19:53:54Z","timestamp":1792439634000},"page":"e101076-e101076","source":"Crossref","is-referenced-by-count":0,"title":["Nonlinear Feature Modeling Using Kolmogorov-Arnold Networks in Electroencephalography for Depression Classification: Deep Learning Study"],"prefix":"10.2196","volume":"5","author":[{"ORCID":"https:\/\/orcid.org\/0009-0000-9202-3493","authenticated-orcid":false,"given":"Bin","family":"Yang","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0001-0046-0125","authenticated-orcid":false,"given":"Yongfeng","family":"Fan","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0003-7180-0654","authenticated-orcid":false,"given":"Xinhong","family":"Wei","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0003-5915-2914","authenticated-orcid":false,"given":"Chunlei","family":"Shi","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1010","published-online":{"date-parts":[[2026,10,19]]},"reference":[{"issue":"3","key":"R1","doi-asserted-by":"publisher","first-page":"539","DOI":"10.22266\/ijies2024.0630.42","article-title":"ACNN-LSTM: a novel deep learning approach for decoding depression from resting-state EEG","volume":"17","author":"Annapoorani","journal-title":"Int J Intell Eng Syst"},{"key":"R2","doi-asserted-by":"publisher","DOI":"10.1051\/matecconf\/202439201101","article-title":"A novel approach for analysis and detection of depression using electroencephalogram (EEG) signals","volume":"392","author":"K","journal-title":"MATEC Web Conf"},{"key":"R3","doi-asserted-by":"crossref","unstructured":"Alneqrat FM Maamar AH Mohammed AN Aghnaya HA . Depression detection based on EEG signals using machine learning algorithms. Presented at: 2024 IEEE 4th International Maghreb Meeting of the Conference on Sciences and Techniques of Automatic Control and Computer Engineering (MI-STA); May 19-21, 2024:625-630; Tripoli, Libya. [doi: 10.1109\/MI-STA61267.2024.10599680]","DOI":"10.1109\/MI-STA61267.2024.10599680"},{"key":"R4","doi-asserted-by":"crossref","unstructured":"Ellis CA Sancho ML Miller RL Calhoun VD . Identifying EEG biomarkers of depression with novel explainable deep learning architectures. bioRxiv. Preprint posted online on Mar 21, 2024. [doi: 10.1101\/2024.03.19.585728] [Medline: 38562835]","DOI":"10.1101\/2024.03.19.585728"},{"key":"R5","doi-asserted-by":"publisher","DOI":"10.1016\/j.eswa.2023.122356","article-title":"A novel EEG-based graph convolution network for depression detection: incorporating secondary subject partitioning and attention mechanism","volume":"239","author":"Zhang","journal-title":"Expert Syst Appl"},{"issue":"1","key":"R6","doi-asserted-by":"publisher","first-page":"763","DOI":"10.1007\/s11042-023-15827-7","article-title":"Analysis of region of interest (RoI) of brain for detection of depression using EEG signal","volume":"83","author":"Mahato","journal-title":"Multimed Tools Appl"},{"key":"R7","doi-asserted-by":"publisher","DOI":"10.1016\/j.bspc.2024.106182","article-title":"EDT: An EEG-based attention model for feature learning and depression recognition","volume":"93","author":"Ying","journal-title":"Biomed Signal Process Control"},{"key":"R8","doi-asserted-by":"publisher","DOI":"10.1016\/j.bspc.2024.106051","article-title":"MGSN: depression EEG lightweight detection based on multiscale DGCN and SNN for multichannel topology","volume":"92","author":"Chen","journal-title":"Biomed Signal Process Control"},{"issue":"9","key":"R9","doi-asserted-by":"publisher","first-page":"5168","DOI":"10.1109\/JBHI.2024.3401389","article-title":"EEGDepressionNet: a novel self attention-based gated DenseNet with hybrid heuristic adopted mental depression detection model using EEG signals","volume":"28","author":"Abidi","journal-title":"IEEE J Biomed Health Inf"},{"issue":"3","key":"R10","doi-asserted-by":"publisher","first-page":"1592","DOI":"10.1109\/TAFFC.2025.3527459","article-title":"TFAGL: a novel agent graph learning method using time-frequency EEG for major depressive disorder detection","volume":"16","author":"Xu","journal-title":"IEEE Trans Affective Comput"},{"key":"R11","doi-asserted-by":"publisher","DOI":"10.1016\/j.jad.2023.11.017","article-title":"Diagnostic deep learning algorithms that use resting EEG to distinguish major depressive disorder, bipolar disorder, and schizophrenia from each other and from healthy volunteers","volume":"346","author":"Ravan","journal-title":"J Affect Disord"},{"key":"R12","unstructured":"Liu Z Wang Y Vaidya S et al. KAN: Kolmogorov-Arnold Networks. arXiv. Preprint posted online on Feb 9, 2025. [doi: 10.48550\/arXiv.2404.19756]"},{"issue":"13","key":"R13","doi-asserted-by":"publisher","first-page":"38819","DOI":"10.1007\/s11042-023-16221-z","article-title":"Hybrid machine learning models to detect signs of depression","volume":"83","author":"Khan","journal-title":"Multimed Tools Appl"},{"issue":"6","key":"R14","doi-asserted-by":"publisher","first-page":"3791","DOI":"10.1007\/s41870-024-01981-0","article-title":"Early depression detection using ensemble machine learning framework","volume":"16","author":"Khan","journal-title":"Int J Inf Tecnol"},{"key":"R15","doi-asserted-by":"crossref","unstructured":"Ayodele A Adetunla A Akinlabi E . RETRACTED: prediction of depression severity and personalised risk factors using machine learning on multimodal data. 2024;20(9):130-143. Retracted in: Int J Online Eng. [doi: 10.3991\/ijoe.v20i09.47581]","DOI":"10.3991\/ijoe.v20i09.47581"},{"key":"R16","doi-asserted-by":"publisher","DOI":"10.1016\/j.neuroimage.2023.120499","article-title":"Prediction of anxious depression using multimodal neuroimaging and machine learning","volume":"285","author":"Zhou","journal-title":"Neuroimage"},{"issue":"16","key":"R17","doi-asserted-by":"publisher","first-page":"49149","DOI":"10.1007\/s11042-023-17369-4","article-title":"An early assessment of persistent depression disorder using machine learning algorithm","volume":"83","author":"Upadhyay","journal-title":"Multimed Tools Appl"},{"issue":"22","key":"R18","doi-asserted-by":"publisher","DOI":"10.3390\/app142210532","article-title":"Impact of feature selection techniques on the performance of machine learning models for depression detection using EEG data","volume":"14","author":"Hassan","journal-title":"Appl Sci"},{"issue":"1-2","key":"R19","doi-asserted-by":"publisher","first-page":"39","DOI":"10.1007\/s11277-023-10445-w","article-title":"A machine learning framework for major depressive disorder (MDD) detection using non-invasive EEG signals","volume":"140","author":"Bashir","journal-title":"Wireless Pers Commun"},{"key":"R20","doi-asserted-by":"publisher","DOI":"10.1016\/j.clinph.2024.05.017","article-title":"EEG based functional connectivity in resting and emotional states may identify major depressive disorder using machine learning","volume":"164","author":"Earl","journal-title":"Clin Neurophysiol"},{"issue":"11","key":"R21","doi-asserted-by":"publisher","DOI":"10.3390\/brainsci14111107","article-title":"Machine-learning-based depression detection model from electroencephalograph (EEG) data obtained by consumer-grade EEG device","volume":"14","author":"Suzuki","journal-title":"Brain Sci"},{"issue":"3","key":"R22","doi-asserted-by":"publisher","DOI":"10.14569\/IJACSA.2024.0150363","article-title":"Hybrid machine learning approaches for predicting and diagnosing major depressive disorder","volume":"15","author":"Balakrishna","journal-title":"Int J Adv Comput Sci Appl (IJACSA)"},{"issue":"3","key":"R23","doi-asserted-by":"publisher","first-page":"581","DOI":"10.35882\/jeeemi.v7i3.719","article-title":"Applied machine learning in EEG data classification to classify major depressive disorder by critical channels","volume":"7","author":"Dhekane","journal-title":"J Electron Electromedical Eng Med Inf"},{"key":"R24","doi-asserted-by":"publisher","DOI":"10.1016\/j.neucom.2025.130008","article-title":"M-MDD: a multi-task deep learning framework for major depressive disorder diagnosis using EEG","volume":"636","author":"Wang","journal-title":"Neurocomputing"},{"issue":"1","key":"R25","doi-asserted-by":"publisher","first-page":"46","DOI":"10.32996\/jcsts.2025.7.1.4","article-title":"Machine learning and deep learning techniques for EEG-based prediction of psychiatric disorders","volume":"7","author":"Abir","journal-title":"J Comput Sci Technol Stud"},{"key":"R26","doi-asserted-by":"publisher","DOI":"10.1080\/23279095.2025.2457999","article-title":"EEG-derived brainwave patterns for depression diagnosis via hybrid machine learning and deep learning frameworks","author":"Ahire","journal-title":"Appl Neuropsychol Adult"},{"key":"R27","doi-asserted-by":"publisher","DOI":"10.1016\/j.bspc.2024.107112","article-title":"A lightweight convolutional transformer neural network for EEG-based depression recognition","volume":"100","author":"Hou","journal-title":"Biomed Signal Process Control"},{"key":"R28","doi-asserted-by":"crossref","unstructured":"Kumar L Kaustubh K Rajkhowa T Mattur SA . EEG-ViLSTM: a deep learning approach for depression detection using EEG signals. Presented at: 2025 National Conference on Communications (NCC); Mar 6-9, 2025:1-6; New Delhi, India. [doi: 10.1109\/NCC63735.2025.10983346]","DOI":"10.1109\/NCC63735.2025.10983346"},{"key":"R29","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2025.3563760","article-title":"Deep learning model for analyzing EEG signal analysis","volume":"13","author":"Gupta","journal-title":"IEEE Access"},{"issue":"1","key":"R30","doi-asserted-by":"publisher","DOI":"10.1007\/s11571-025-10305-2","article-title":"Novel EEG-based diagnostic framework for major depressive disorder using microstate and entropy features","volume":"19","author":"Rahmati","journal-title":"Cognit Neurodynamics"},{"key":"R31","doi-asserted-by":"publisher","DOI":"10.1016\/j.bspc.2024.106973","article-title":"A graph-based method for automatic graded diagnosis of depression using EEG signals","volume":"100","author":"Ren","journal-title":"Biomed Signal Process Control"},{"key":"R32","doi-asserted-by":"publisher","DOI":"10.1016\/j.bspc.2024.106930","article-title":"Depression detection based on the temporal-spatial-frequency feature fusion of EEG","volume":"100","author":"Xi","journal-title":"Biomed Signal Process Control"},{"issue":"2","key":"R33","doi-asserted-by":"publisher","DOI":"10.3390\/bioengineering12020095","article-title":"TSF-MDD: a deep learning approach for electroencephalography-based diagnosis of major depressive disorder with temporal-spatial-frequency feature fusion","volume":"12","author":"Gan","journal-title":"Bioengineering (Basel)"},{"key":"R34","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2025.3552619","article-title":"Identification of depression patients using LIF spiking neural network model from the pattern of EEG signals","volume":"13","author":"Sahu","journal-title":"IEEE Access"},{"issue":"11","key":"R35","doi-asserted-by":"publisher","first-page":"12640","DOI":"10.1109\/TMC.2025.3586591","article-title":"LSNN model: a lightweight spiking neural network-based depression classification model for wearable EEG sensors","volume":"24","author":"Zhao","journal-title":"IEEE Trans Mobile Comput"},{"key":"R36","doi-asserted-by":"crossref","unstructured":"M S Priya SJ Subathra MSP A S George ST . Classification of major depressive disorder based on EEG signals using superlet transformation and ResNet-18. Presented at: 2025 8th International Conference on Trends in Electronics and Informatics (ICOEI); Apr 24-26, 2025:1538-1545; Tirunelveli, India. [doi: 10.1109\/ICOEI65986.2025.11013598]","DOI":"10.1109\/ICOEI65986.2025.11013598"},{"issue":"3","key":"R37","doi-asserted-by":"publisher","first-page":"2130","DOI":"10.1109\/TAFFC.2025.3555433","article-title":"UA-DAAN: an uncertainty-aware dynamic adversarial adaptation network for EEG-based depression recognition","volume":"16","author":"Shen","journal-title":"IEEE Trans Affective Comput"},{"key":"R38","doi-asserted-by":"crossref","unstructured":"Deng X Liang W Zou M Wang H Li Y . AttenStateNet: depression classification method for EEG signals based on state space and attention mechanisms. Presented at: 2025 IEEE 14th Data Driven Control and Learning Systems (DDCLS); May 9-11, 2025; Wuxi, China. [doi: 10.1109\/DDCLS66240.2025.11065508]","DOI":"10.1109\/DDCLS66240.2025.11065508"},{"key":"R39","doi-asserted-by":"publisher","DOI":"10.1016\/j.patcog.2025.112478","article-title":"Distribution entropy regularized multimodal subspace support vector data description for anomaly detection","volume":"172","author":"Wang","journal-title":"Pattern Recognit"},{"key":"R40","doi-asserted-by":"publisher","DOI":"10.1016\/j.cropro.2024.106824","article-title":"Knowledge correction and \u03b5-insensitive criterion-leveraged zero-order TSK fuzzy system for rice leaf disease diagnosis[J]","volume":"184","author":"Wang","journal-title":"Crop Protection"},{"issue":"4","key":"R41","doi-asserted-by":"publisher","first-page":"1349","DOI":"10.1109\/TFUZZ.2026.3658901","article-title":"Knowledge calibration fusion and label space graph regularization-based multicenter fuzzy systems","volume":"34","author":"Wang","journal-title":"IEEE Trans Fuzzy Syst"},{"key":"R42","doi-asserted-by":"crossref","unstructured":"Zhu X Qian P Wang C et al. MMFNet: a multi-modal and multi-attention fusion network for multi-task skin disease classification. Presented at: 2024 IEEE International Conference on Bioinformatics and Biomedicine (BIBM); Dec 3-6, 2024:1779-1782; Lisbon, Portugal. [doi: 10.1109\/BIBM62325.2024.10822724]","DOI":"10.1109\/BIBM62325.2024.10822724"},{"issue":"10","key":"R43","doi-asserted-by":"publisher","first-page":"17386","DOI":"10.1109\/TITS.2024.3478051","article-title":"Multi-source patch feature fusion with neighborhood flash attention transformer for pixel-level vehicle and road recognition in hyperspectral image","volume":"26","author":"Cai","journal-title":"IEEE Trans Intell Transport Syst"},{"key":"R44","doi-asserted-by":"publisher","DOI":"10.1016\/j.compbiomed.2023.106782","article-title":"A gated temporal-separable attention network for EEG-based depression recognition","volume":"157","author":"Yang","journal-title":"Comput Biol Med"},{"issue":"4","key":"R45","doi-asserted-by":"publisher","first-page":"1985","DOI":"10.1109\/TAFFC.2024.3392904","article-title":"MAST-GCN: multi-scale adaptive spatial-temporal graph convolutional network for EEG-based depression recognition","volume":"15","author":"Lu","journal-title":"IEEE Trans Affective Comput"},{"issue":"1","key":"R46","doi-asserted-by":"publisher","first-page":"3392","DOI":"10.1109\/TCE.2024.3370310","article-title":"A depression detection auxiliary decision system based on multi-modal feature-level fusion of EEG and speech","volume":"70","author":"Ning","journal-title":"IEEE Trans Consum Electron"},{"issue":"4","key":"R47","doi-asserted-by":"publisher","first-page":"1088","DOI":"10.1038\/s41380-023-02395-3","article-title":"Resting-state EEG dynamic functional connectivity distinguishes non-psychotic major depression, psychotic major depression and schizophrenia","volume":"29","author":"Chen","journal-title":"Mol Psychiatry"},{"issue":"1","key":"R48","doi-asserted-by":"publisher","first-page":"152","DOI":"10.1109\/JBHI.2024.3487012","article-title":"An on-board executable multi-feature transfer-enhanced fusion model for three-lead EEG sensor-assisted depression diagnosis","volume":"29","author":"Tian","journal-title":"IEEE J Biomed Health Inf"},{"issue":"9","key":"R49","doi-asserted-by":"publisher","first-page":"5247","DOI":"10.1109\/JBHI.2024.3404664","article-title":"HEMAsNet: a hemisphere asymmetry network inspired by the brain for depression recognition from electroencephalogram signals","volume":"28","author":"Shen","journal-title":"IEEE J Biomed Health Inf"},{"key":"R50","doi-asserted-by":"publisher","DOI":"10.1016\/j.jad.2023.10.157","article-title":"Altered resting-state brain oscillation and the associated cognitive impairments in late-life depression with different depressive severity: an EEG power spectrum and functional connectivity study","volume":"348","author":"Zeng","journal-title":"J Affect Disord"},{"issue":"3","key":"R51","doi-asserted-by":"publisher","first-page":"804","DOI":"10.1093\/schbul\/sbae150","article-title":"EEG-based signatures of schizophrenia, depression, and aberrant aging: a supervised machine learning investigation","volume":"51","author":"Sarisik","journal-title":"Schizophr Bull"},{"key":"R52","doi-asserted-by":"publisher","DOI":"10.1016\/j.jad.2024.09.054","article-title":"Enhanced network synchronization connectivity following transcranial direct current stimulation (tDCS) in bipolar depression: effects on EEG oscillations and deep learning-based predictors of clinical remission","volume":"369","author":"Xiao","journal-title":"J Affect Disord"},{"key":"R53","doi-asserted-by":"publisher","DOI":"10.1016\/j.bspc.2023.105003","article-title":"Functional connectivity and power spectral density analysis of EEG signals in trained practitioners of Bhramari Pranayama","volume":"84","author":"Malan","journal-title":"Biomed Signal Process Control"},{"key":"R54","doi-asserted-by":"publisher","DOI":"10.1016\/j.neucom.2023.126387","article-title":"Novel welch-transform based enhanced spectro-temporal analysis for cognitive microsleep detection using a single electrode EEG","volume":"549","author":"Shah","journal-title":"Neurocomputing"},{"issue":"1","key":"R55","doi-asserted-by":"publisher","DOI":"10.1080\/19942060.2024.2322514","article-title":"Two-phase flow investigation of sewage pumps with different tip clearance via computational fluid dynamics and multi-factor ANOVA","volume":"18","author":"Yang","journal-title":"Eng Appl Comput Fluid Mech"},{"key":"R56","doi-asserted-by":"publisher","DOI":"10.1016\/j.eswa.2023.121658","article-title":"An EEG-based cross-subject interpretable CNN for game player expertise level classification","volume":"237","author":"Lin","journal-title":"Expert Syst Appl"},{"issue":"7","key":"R57","doi-asserted-by":"publisher","first-page":"2420","DOI":"10.1007\/s11263-024-01989-w","article-title":"Oriented R-CNN and beyond","volume":"132","author":"Xie","journal-title":"Int J Comput Vis"},{"key":"R58","doi-asserted-by":"publisher","DOI":"10.1016\/j.energy.2024.130326","article-title":"Fault diagnosis of hydro-turbine via the incorporation of Bayesian algorithm optimized CNN-LSTM neural network","volume":"290","author":"Dao","journal-title":"Energy"},{"key":"R59","doi-asserted-by":"publisher","DOI":"10.1016\/j.compbiomed.2023.107758","article-title":"DM-CNN: dynamic multi-scale convolutional neural network with uncertainty quantification for medical image classification","volume":"168","author":"Han","journal-title":"Comput Biol Med"},{"key":"R60","doi-asserted-by":"publisher","DOI":"10.1016\/j.neucom.2024.127284","article-title":"An optimized CNN-BiLSTM network for bearing fault diagnosis under multiple working conditions with limited training samples","volume":"574","author":"Song","journal-title":"Neurocomputing"},{"issue":"21","key":"R61","doi-asserted-by":"publisher","DOI":"10.3390\/rs16214015","article-title":"How to learn more? Exploring Kolmogorov\u2013Arnold networks for hyperspectral image classification","volume":"16","author":"Jamali","journal-title":"Remote Sens (Basel)"}],"container-title":["JMIR AI"],"original-title":[],"language":"en","deposited":{"date-parts":[[2026,10,19]],"date-time":"2026-10-19T19:53:56Z","timestamp":1792439636000},"score":1,"resource":{"primary":{"URL":"https:\/\/ai.jmir.org\/2026\/1\/e101076"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,10,19]]},"references-count":61,"URL":"https:\/\/doi.org\/10.2196\/101076","relation":{},"ISSN":["2817-1705"],"issn-type":[{"value":"2817-1705","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,10,19]]},"article-number":"v5i3e101076"}}