Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

"Understanding" medical school curriculum content using KnowledgeMap.

Joshua C Denny1, Jeffrey D Smithers, Randolph A Miller

  • 1Department of Biomedical Informatics, Vanderbilt School of Medicine, Nashville, Tennessee 37232, USA.

Journal of the American Medical Informatics Association : JAMIA
|April 2, 2003
PubMed
Summary

The KnowledgeMap (KM) tool accurately identifies biomedical concepts in medical curricula, outperforming MetaMap. This computational approach aids in curriculum development and review.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Systematic common and rare variant association testing in 392,030 whole genomes in <i>All of Us</i>.

medRxiv : the preprint server for health sciences·2026
Same author

Multi-scale data improves performance of machine learning model for long COVID identification.

Communications medicine·2026
Same author

Downward bias in the association between APOE and Alzheimer's disease using prevalent and by-proxy disease sampling in the All of Us research program.

BMC medical genomics·2026
Same author

Genetic Prediction of Circulating Lipoprotein(a) Levels in Diverse Populations.

medRxiv : the preprint server for health sciences·2026
Same author

11 million days of longitudinal wearable data reveal novel future health insights.

medRxiv : the preprint server for health sciences·2026
Same author

A one-shot, lossless algorithm for cross-cohort learning in mixed-outcomes analysis.

Patterns (New York, N.Y.)·2025

Area of Science:

  • Medical Informatics
  • Computational Linguistics
  • Biomedical Education

Background:

  • Medical school curricula require robust tools for concept identification and integration.
  • Existing tools may not fully capture the complexity of biomedical terminology within educational documents.

Purpose of the Study:

  • To develop and evaluate computational tools for identifying concepts in medical curricular documents.
  • To leverage the National Library of Medicine's Unified Medical Language System (UMLS) for concept extraction.
  • To enhance faculty and student abilities in curriculum development, review, and integration.

Main Methods:

  • Developed the KnowledgeMap (KM) concept identifier using UMLS resources, heuristic language processing, and an empirical scoring algorithm.
  • Compared KM's concept recognition against the National Library of Medicine's MetaMap program.

Related Experiment Videos

  • Manually identified "gold standard" biomedical concepts in medical lecture documents for evaluation.

    • Main Results:

    • KM achieved 82% concept matching and 89% precision, exceeding MetaMap's 78% match rate and 85% precision.
    • KM effectively identified acronyms, underspecified concepts, and ambiguous matches.
    • The primary limitation was concepts not present in the UMLS Metathesaurus.

    Conclusions:

    • The prototype KM system demonstrates promising concept extraction capabilities for medical curricula.
    • Future KM versions could facilitate curriculum navigation, identify redundancies, and pinpoint omissions.
    • Assessing KM's utility for personalized information needs is recommended.