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

Text categorization models for high-quality article retrieval in internal medicine.

Yindalon Aphinyanaphongs1, Ioannis Tsamardinos, Alexander Statnikov

  • 1Department of Biomedical Informatics, 4th Floor, Eskind Biomedical Library, 2209 Garland Avenue, Vanderbilt University, Nashville, TN 37232, USA. ping.pong@vanderbilt.edu

Journal of the American Medical Informatics Association : JAMIA
|November 25, 2004
PubMed
Summary

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

General-purpose large language models outperform specialized clinical AI tools on medical benchmarks.

Nature medicine·2026
Same author

A Lasting Legacy: Long-Term Effects of Exercise Training on Cardiometabolic Health in the STRRIDE-Prediabetes Reunion Study.

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

Theory and practice in biomedical informatics: a framework for discovery.

Journal of the American Medical Informatics Association : JAMIA·2026
Same author

Regulation of Small RNAs by Exercise and Their Role in Insulin Sensitivity.

bioRxiv : the preprint server for biology·2026
Same author

CNS-Obsidian: A Neurosurgical Vision-Language Model Built From Scientific Publications.

Neurosurgery·2026
Same author

Generalist Foundation Models Are Not Clinical Enough for Hospital Operations.

Research square·2026

Machine learning models can automatically identify high-quality internal medicine articles, outperforming older PubMed filters. This improves evidence retrieval for clinical decision-making.

Area of Science:

  • Medical informatics
  • Machine learning in medicine
  • Evidence-based medicine

Background:

  • The exponential growth of medical literature makes identifying high-quality, relevant scientific evidence challenging for clinicians.
  • Existing Boolean-based PubMed clinical query filters may not be optimal for retrieving specific, high-quality articles.

Purpose of the Study:

  • To apply machine learning techniques for automatically identifying high-quality, content-specific internal medicine articles.
  • To compare the performance of machine learning models against traditional PubMed clinical query filters.

Main Methods:

  • Machine learning algorithms including Naive Bayes, AdaBoost, and Support Vector Machines (SVM) were trained.
  • Article selection criteria from ACP Journal Club served as the gold standard for high-quality articles.

Related Experiment Videos

  • Performance was evaluated using area under the receiver operating characteristic curves (AUC) and 11-point average recall precision.

    • Main Results:

    • Machine learning models demonstrated superior or comparable sensitivity, specificity, and precision compared to clinical query filters.
    • Polynomial SVM models achieved the best performance in ranking articles, as indicated by AUC and precision metrics.

    Conclusions:

    • Machine learning methods can effectively automate the retrieval of high-quality, content-specific medical articles.
    • These data-driven models show improved performance over the 1994 PubMed clinical query filters for internal medicine evidence retrieval.