방사선과 보고서의 효율적인 UMLS concept indexing을 위한 연구

이정애1), 서화정1), 김기원1), 김민구1), 홍승권1), 박유랑1), 김주한*1)

1)서울대학교 의과대학 생명의료정보연구실

Abstract : For the effective retrieval of clinical information, the indexing process is essential. We evaluated the UMLS subset mapping and semantic type filtering methods to improve the precision of UMLS concept indexing in diagnosis section of radiology reports by MetaMap. In this study MSH2004+RCD99+SNMI+UWDA and ICD99+SNMI+UWDA vocabulary combinations are useful for the improvement in precision. After calculating the candidates from subset combinations, we obtained more effectively improved results by semantic type filtering.
keyword : UMLS concept indexing, UMLS subset mapping, Semantic type filtering

연제분류

,


첨부한 파일

Kosmi_jalee.hwp

첨부한 e-poster파일

투고내용을 삭제하시려면 "삭제"버튼을 누르시고 다시 접수하여 주시기 바랍니다.

 암 호

대한의료정보학회
Copyright (c) 2002 by The Korean Society of Medical Informatics