Application of a Medical Text Indexer to an Online Dermatology Atlas

George R. Kim, A. R. Aronson, J. G. Mork, B. A. Cohen, C. U. Lehmann

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Clinical dermatology cases are presented as images and semi-structured text describing skin lesions and their relationships to disease. Metadata assignment to such cases is hampered by lack of a standardized dermatology vocabulary and facilitated methods for indexing legacy collections. In this pilot study descriptive clinical text from Dermatlas, a Web-based repository of dermatology cases, was indexed to Medical Subject Heading (MeSH®) terms using the National Library of Medicine's Medical Text Indexer (MTI). The MTI is an automated text processing system that derives ranked lists of MeSH terms to describe the content of medical journal citations using knowledge from the Unified Medical Language System® (UMLS®) and from MEDLINE®. For a representative, random sample of 50 Dermatlas cases, the MTI frequently derived MeSH indexing terms that matched expert-assigned terms for Diagnoses (88%), Lesion Types (72%), and Patient Characteristics (Gender and Age Groups, 62% and 84% respectively). This pilot demonstrates the potential for extending the MTI to automate indexing of clinical case presentations and for using MeSH to describe aspects of clinical dermatology.

Original languageEnglish (US)
Title of host publicationStudies in Health Technology and Informatics
EditorsMarius Fieschi, Enrico W. Coiera, Yu-Chuan Li
PublisherIOS Press BV
Pages287-291
Number of pages5
ISBN (Electronic)1586034448, 9781586034443
StatePublished - 2004
Event11th World Congress on Medical Informatics, MEDINFO 2004 - San Francisco, United States
Duration: Sep 7 2004Sep 11 2004

Publication series

NameStudies in Health Technology and Informatics
Volume107
ISSN (Print)0926-9630
ISSN (Electronic)1879-8365

Conference

Conference11th World Congress on Medical Informatics, MEDINFO 2004
Country/TerritoryUnited States
CitySan Francisco
Period9/7/049/11/04

Keywords

  • Controlled Vocabulary
  • Dermatology
  • Medical Informatics
  • Natural Language Processing

ASJC Scopus subject areas

  • Biomedical Engineering
  • Health Informatics
  • Health Information Management

Fingerprint

Dive into the research topics of 'Application of a Medical Text Indexer to an Online Dermatology Atlas'. Together they form a unique fingerprint.

Cite this