Shorten Bipolarity Checklist for the Differentiation of Subtypes of Bipolar Disorder Using Machine Learning

摘要：

　　The differentiation of type Ⅰ and type Ⅱ of bipolar disorder is difficult.In clinical practices, corresponding diagnostic operability is poor since their criterions are similar and do not include past or lifetime characteristics.The aim of this study was to generate the clinical feasible scale by using machine learning algorithms based on the analysis of a Chinese multi-center cohort data.To evaluate the importance of each item of Affective Disorder Evaluation(ADE), a case-control study of Chinese samples including 281 type Ⅰ of bipolar disorder and 79 type Ⅱ of bipolar disorder patients conducted from 9 Chinese health facilities participating in CAF(E)-BD.The novel scale was formed by selected items from ADE according to its importance calculated by mutual information criteria of minimal-redundancy-maximal-relevance(mRMR).

关键词： Bipolar disorder subtype ADE machine learning algorithm mRMR

作者: Chaonan Feng Huimin Gao Xuefeng B Ling Jun Ji Yantao Ma

作者单位: College of Computer Science and Technology Qingdao University China Peking University Institute of Mental Health Beijing China Department of Surgery, Stanford University California United States of America

会议类型: 国际会议

会议名称: 2018 6th International Conference on Bioinformatics and Computational Biology(ICBCB 2018)(第六届生物信息学与计算生物学国际会议)

会议地点: 成都

会议语种:英文

页码: 162-166

在线出版日期: 2018-03-12（万方平台首次上网日期，不代表论文的发表时间）