Using UMLS metathesaurus concepts to describe medical images: dermatology vocabulary

Authors:
James W. Woods;Charles A. Sneiderman;Kamran Hameed;Michael J. Ackerman;Charlie Hatton
Affiliations:
National Library of Medicine, Ackerman-Office of High Performance Computing and Communications, 38A/B1N30F1, Bethesda, MD 20894, USA;National Library of Medicine, Ackerman-Office of High Performance Computing and Communications, 38A/B1N30F1, Bethesda, MD 20894, USA;National Library of Medicine, Department of Medicine, Office of High Performance Computing and Communications, The Aga Khan University, Karachi, Pakistan;National Library of Medicine, Ackerman-Office of High Performance Computing and Communications, 38A/B1N30F1, Bethesda, MD 20894, USA;Brookline, MA, USA
Venue:
Computers in Biology and Medicine
Year:
2006

Citing 0
Cited 4

MPEG-7 as a Metadata Standard for Indexing of Surgery Videos in Medical E-Learning

ICCS '08 Proceedings of the 8th international conference on Computational Science, Part III
Associating Clinical Archetypes Through UMLS Metathesaurus Term Clusters

Journal of Medical Systems
Linking uncertainty in physicians' narratives to diagnostic correctness

ExProM '12 Proceedings of the Workshop on Extra-Propositional Aspects of Meaning in Computational Linguistics
Annotation schemes to encode domain knowledge in medical narratives

LAW VI '12 Proceedings of the Sixth Linguistic Annotation Workshop

Quantified Score

Hi-index	0.00

Visualization

Abstract

Web servers at the National Library of Medicine (NLM) displayed images of ten skin lesions to practicing dermatologists and provided an online form for capturing text they used to describe the pictures. The terms were submitted to the UMLS Metathesaurus (Meta). Concepts retrieved, their semantic types, definitions and synonyms, were returned to each subject in a second web-based form. Subjects rated the concepts against their own descriptive terms. They submitted 825 terms, 346 of which were unique and 300 mapped to UMLS concepts. The dermatologists rated 295 concepts as 'Exact Match' and they accomplished both tasks in about 30min.