基于脑结构像的精神分裂症机器学习分类  

The classification of schizophrenia based on brain structural features:A machine learning approach

作　　者：郑泓 蒲城城[3] 王毅[1,2] 陈楚侨 ZHENG Hong;PU Cheng-cheng;WANG Yi;Raymond CKCHAN(Neuropsychology and Applied Cognitive Neuroscience Laboratory,CAS Key Laboratory of Mental Health,Institute of Psychology,Chinese Academy of Sciences,Beijing 100101,China;Department of Psychology,University of Chinese Academy of Sciences,Beijing 100049,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 100191,China)

机构地区：[1]中国科学院心理研究所心理健康重点实验室,神经心理学与认知神经科学研究室,北京100101 [2]中国科学院大学心理系,北京100049 [3]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),北京100191

出　　处：《心理科学进展》2020年第2期252-265,共14页Advances in Psychological Science

基　　金：国家自然科学基金项目(31871114,31400884)

摘　　要：将机器学习应用于精神疾患的临床和基础研究是近年来的趋势。研究者将机器学习应用于精神分裂症患者及高危人群的T1加权像和弥散张量成像的脑影像数据中,为了解疾病的生理病理学机制提供帮助。回顾以往研究发现额叶及颞叶的脑结构特征具有较高的区分能力,行为数据和脑影像数据结合的分类效果优于单模态数据。现阶段研究存在样本量不足和泛化能力欠缺的局限,未来研究应注意扩大样本量、制定标准化的分类方法,从而进一步探究机器学习在精神疾患中的作用。Machine learning is a promising approach for mental disorders.In recent years,machine learning based on T1 weighted imaging and Diffusion Tensor Imaging(DTI)data has been used to investigate the psychopathology and underlying mechanisms of schizophrenia patients and high-risk population.The findings from the previous literature suggest that structural features of frontal lobe and temporal lobe can improve classification performance.In addition,the combination of behavioural performances and the features of brain structure is superior to the single-modality structural images on classification accuracy.However,the existing empirical studies classifying schizophrenia patients or high-risk population from controls are limited in sample size and generalization ability.

关 键 词：脑结构像 弥散张量成像 机器学习 精神分裂症 高危人群 

分 类 号：R395[哲学宗教—心理学]