Multi-lingual features of the Unified Medical Language System.

Bodenreider O

Proc. CLEF-ER 2013 workshop 2013.


The Unified Medical Language System (UMLS) is a terminology integration system developed and maintained by the U.S. National Library of Medicine (NLM). Over the past 20 years, the UMLS Metathesaurus has been extended to encompass 168 source vocabularies. While English is the dominant language (116 source vocabularies), 52 vocabularies in the Metathesaurus are in languages other than English (6 in Dutch, German and Spanish ; 5 in French ; 4 in Italian and Portuguese ; 2 in Czech, Finnish, Hungarian, Japanese, Korean, Norwegian and Swedish ; and 1 in Basque, Croatian, Danish, Hebrew, Latvian, Polish and Russian).NLM does not translate any of these vocabularies, but rather integrates translations of English vocabularies when they are available. Examples include the Medical Subject Headings (MeSH) present in the Metathesaurus in 16 languages ; the International Classification of Primary Care (ICPC) in 14 ; the Medical Dictionary for Regulatory Activities Terminology (MedDRA) in 10 ; and the WHO Adverse Drug Reaction Terminology in 5. Of note, SNOMED CT and the Physicians' Current Procedural Terminology (CPT) are both present in English and Spanish, because these two languages are the main languages used in the U.S.

Bodenreider O. Multi-lingual features of the Unified Medical Language System. 
Proc. CLEF-ER 2013 workshop 2013.