脑电信号在抑郁和焦虑障碍识别中的应用  

Application of electroencephalogram for depression and anxiety identification

作　　者：荆舒 戴振威 苏小游[1] Jing Shu;Dai Zhenwei;Su Xiaoyou(School of Population Medicine and Public Health,Chinese Academy of Medical Sciences&Peking Union Medical College,Beijing 100010,China;Peking University Sixth Hospital,Peking University Institute of Mental Health,NHC Key Laboratory of Mental Health(Peking University),National Clinical Research Center for Mental Disorders(Peking University Sixth Hospital),Beijing 100080,China)

机构地区：[1]中国医学科学院北京协和医学院群医学及公共卫生学院,北京100010 [2]北京大学第六医院,北京大学精神卫生研究所,国家卫生健康委员会精神卫生学重点实验室(北京大学),国家精神心理疾病临床医学研究中心(北京大学第六医院),北京100080

出　　处：《中华行为医学与脑科学杂志》2025年第1期89-94,共6页Chinese Journal of Behavioral Medicine and Brain Science

基　　金：中国医学科学院医学与健康科技创新工程项目(CAMS 2021-I2M-1-004)。

摘　　要：抑郁和焦虑是最常见的两类精神障碍,会对个体及社会造成严重的负面影响,早期、准确地识别抑郁和焦虑障碍对患者的症状控制和预后具有重要意义。传统的识别方式,如量表筛查、访谈等均依赖于患者的自我回答及精神科医生的综合判断,存在主观性、依赖于医疗资源可及性和便利性等固有局限,因此寻找客观手段辅助抑郁和焦虑障碍的识别已成为近年来的研究重点。脑电信号是对头皮表面电势差变化的描述,具有客观、量化、时间分辨率高等优势。脑电信号已成为当前识别抑郁和焦虑障碍的潜在客观指标之一。机器学习算法是提取脑电信号特征、提高识别准确率的关键技术,基于EEG和机器学习算法的自动化诊断有望成为抑郁和焦虑障碍识别的新手段。此外,便携式脑电采集为快速识别精神障碍和开展大规模筛查行动提供了可能。本文综述了脑电信号在抑郁和焦虑障碍识别中的应用情况,供相关领域学者参考和借鉴,以期推动我国精神障碍识别的客观化、可视化发展。Depression and anxiety are the two most common mental disorders,which have severe negative impacts on individuals and society.Early and accurate identification of these disorders is crucial for symptom control and positive prognosis.Traditional identification approaches such as scale screening and interviews rely on patients'self-reporting and the comprehensive judgment of psychiatrists.These methods inherently have limitations such as subjectivity and dependence on the availability and convenience of medical resources.Therefore,seeking objective means to assist in the identification of depression and anxiety disorders has become a research focus in recent years.Electroencephalogram(EEG),which describes changes in potential difference on the scalp surface,offers advantages such as objectivity,quantification and high temporal resolution.EEG has become one of the potential objective indicators for identifying depression and anxiety disorders.Machine learning algorithms are key technologies for extracting EEG signal features and improving identification accuracy.The combination of EEG collection and machine learning algorithms is expected to become a new method for identifying depression and anxiety disorders.In addition,portable EEG collection provides the possibility for rapid identification of mental disorders and large-scale screening activities.This article reviews the application of EEG for depression and anxiety disorders identification,providing a reference for scholars in related fields,with the aim of promoting the development of mental disorder identification towards objectivity and visualization in China.

关 键 词：脑电图 抑郁障碍 焦虑障碍 机器学习 

分 类 号：TN911.7[电子电信—通信与信息系统] R749.4[电子电信—信息与通信工程] R749.72[医药卫生—神经病学与精神病学]