Fung Kin Wah, Bodenreider Olivier
National Library of Medicine, Bethesda, Maryland, USA. {kwfung|olivier}@nlm.nih.gov
AMIA Annu Symp Proc. 2005;2005:266-70.
An algorithm was derived to find candidate mappings between any two terminologies inside the UMLS, making use of synonymy, explicit mapping relations and hierarchical relationships among UMLS concepts. Using an existing set of mappings from SNOMED CT to ICD9CM as our gold standard, we managed to find candidate mappings for 86% of SNOMED CT terms, with recall of 42% and precision of 20%. Among the various methods used, mapping by UMLS synonymy was particularly accurate and could potentially be useful as a quality assurance tool in the creation of mapping sets or in the UMLS editing process. Other strengths and weaknesses of the algorithm are discussed.
我们推导了一种算法,利用统一医学语言系统(UMLS)中概念间的同义关系、显式映射关系和层次关系,来寻找UMLS内任意两个术语之间的候选映射。以从医学系统命名法临床术语(SNOMED CT)到国际疾病分类第九版临床修订本(ICD9CM)的现有映射集作为金标准,我们成功为86%的SNOMED CT术语找到了候选映射,召回率为42%,精确率为20%。在使用的各种方法中,通过UMLS同义词进行映射特别准确,在创建映射集或UMLS编辑过程中,有可能作为一种质量保证工具。我们还讨论了该算法的其他优缺点。