机器学习在抑郁症辅助诊断中的应用进展  

Advances in the application of machine learning in auxiliary diagnosis of depression

作　　者：刁云恒 王慧颖 董娇[1] 朱艺菡 邵秋静 冯来鹏 王长虹[1,2,4] DIAO Yunheng;WANG Huiying;DONG Jiao;ZHU Yihan;SHAO Qiujing;FENG Laipeng;WANG Changhong(The Second Affiliated Hospital of Xinxiang Medical University(He'nan Mental Hospital),Xinxiang 453002,China;He'nan Key Laboratory of Biological Psychiatry,Xinxiang Medical University,Xinxiang 453002,China;The Second Clinical College,Xinxiang Medical University,Xinxiang 453002,China;He'nan Provincial Psychological Assistance Cloud Platform and Application Engineering Research Center,Xinxiang 453002,China)

机构地区：[1]新乡医学院第二附属医院(河南省精神病医院),河南新乡453002 [2]新乡医学院河南省生物精神病学重点实验室,河南新乡453002 [3]新乡医学院第二临床学院,河南新乡453002 [4]河南省心理援助云平台及应用工程研究中心,河南新乡453002

出　　处：《中国医学物理学杂志》2022年第2期257-264,共8页Chinese Journal of Medical Physics

基　　金：新乡市重大科技专项(ZD2020011);河南省医学科技攻关计划项目(2018010015,2018020375);河南省精神心理疾病临床医学研究中心开放课题(2020-zxkfkt-004);大学生创新创业训练计划项目(202110472030)。

摘　　要：随着抑郁症诊疗技术的发展,各种抑郁症相关的临床数据量急速扩增,机器学习技术恰好适用于大数量、多维度、多模态的数据,通过机器学习技术自动学习抑郁症诊疗数据中的特征,利用数据特征对抑郁症进行疾病诊断、疗效预测,达到抑郁症辅助诊断的目的。本文从机器学习在不同种类临床数据上应用的角度对文献进行了系统性分析,总结了机器学习在抑郁症辅助诊断领域的通用研究流程及常用研究方法,并展望未来的研究方向以及面临的挑战。With the development of technologies in the diagnosis and treatment of depression,the volume of clinical data related to depression is rapidly expanding,and machine learning technology is just suitable for large quantity,multi-dimensional,multimode data.The characteristics of the data obtained in the diagnosis and treatment of depression can be automatically learned by machine learning technology,and the data features can be used for the diagnosis of depression and the prediction of curative effect,so as to achieve the purpose of auxiliary diagnosis of depression.Herein a systematic analysis is carried out on the literatures about the application of machine learning on various categories of clinical data,and the research process and methods of machine learning in the diagnosis of depression are summarized,and finally,the research directions and challenges in the future are presented.

关 键 词：抑郁症 机器学习 辅助诊断 综述 

分 类 号：R318[医药卫生—生物医学工程] R749.41[医药卫生—基础医学]