重庆理工大学学报2025，Vol.39Issue(9)：73-81,9.DOI:10.3969/j.issn.1674-8425(z).2025.05.009
稀疏贝叶斯学习在脑电抑郁症识别中的应用研究
Research on the application of sparse bayesian learning in EEG-based depression recognition
摘要
Abstract
As mental health issues become increasingly prominent,the high incidence of depression,especially in the post-pandemic era,has garnered widespread attention.Traditional diagnosis of depression relies on the judgment of clinicians,which has certain limitations.Therefore,developing an objective and accurate automatic recognition method for depression is of paramount importance.This study aims to analyze Electroencephalography(EEG)signals to develop an automatic recognition model for depression,thereby enhancing the objectivity and accuracy of diagnoses.Based on the MODMA dataset specifically designed for the analysis of psychological disorders,including the resting-state EEG data from 24 subjects with depression and 29 subjects without,a sparse Bayesian learning(SBL)algorithm is employed to develop an end-to-end depression recognition model.By conducting an in-depth analysis of EEG signals,the differences in brain activity between subjects with depression and those without are explored.Experimental results demonstrate the proposed model achieves 100%in accuracy on the testset,significantly surpassing existing depression detection techniques.Parameter analysis further confirms the model's effectiveness and its value,providing a new perspective for the automated detection and diagnosis of depression.关键词
抑郁症识别/脑电信号/客观评估/稀疏贝叶斯学习Key words
depression recognition/EEG signals/objective assessment/SBL分类
信息技术与安全科学引用本文复制引用
沈如达,朱洁,苏吉普,赵焱,何万源,常洪丽..稀疏贝叶斯学习在脑电抑郁症识别中的应用研究[J].重庆理工大学学报,2025,39(9):73-81,9.基金项目
国家自然科学基金项目(62206210) （62206210）
