摘要
针对国内电子病历(EMR)中临床日常用语特点,通过字面及语义相似度等完成了部分电子病历中的临床日常用语与人类表型术语集术语概念的映射,这对主流医学术语集富集中文电子病历同义词有较大的实践意义。研究结果显示,通过相似度算法和人工验证相结合,可有效降低中文同义词富集的人力消耗。随着基于众包机制的术语维护,电子病历中日常用语可被持续不断地和低成本地富集到人类表型术语集等主流医学术语集上。
Certain daily used clinical terms in electronic medical records(EMR)and the concept of human phenotype terms were mapped according to the literal and semantic similarity based on the characteristics of daily used clinical terms in EMR,which is of great significance for enriching the Chinese synonyms in major medical terms.The study showed that the Chinese synonyms enriched using the similarity algorithm combined with artificial verification can effectively reduce the drain on manpower.The daily used clinical terms in EMR can be continuously enriched to the major medical terms such as human phenotype terms at a low cost with the maintenance of terms based on the mechanism of crowd sourcing.
作者
张睿
陈薇
杨豪
刘然
王觅也
ZHANG Rui;CHEN Wei;YANG Hao;LIU Ran;WANG Mi-ye(Sichuan University West China Hospital Information Center, Chengdu 610041, Sichuan Province, China;Sichuan University West China Hospital Information Center and Information Technology Engineering Research Center, Chengdu 610041,Sichuan Province, China;Sichuan University West China Nursing Innovation Research Center, Chengdu 610041,Sichuan Province, China;Department of Pharmacy, Sichuan Maternity and Infant Health Institute, Chengdu 610045,Sichuan Province, China)
出处
《中华医学图书情报杂志》
CAS
2021年第2期25-32,共8页
Chinese Journal of Medical Library and Information Science
基金
四川省科技厅重点研发项目“基于表型术语的罕见病早期预警决策支持系统构建”(2019YFS0152)。