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

Automated encoding of clinical documents based on natural language processing.

Carol Friedman1, Lyudmila Shagina, Yves Lussier

  • 1Department of Biomedical Informatics, Columbia University, 622 West 168 Street, VC-5, New York, NY 10032, USA. friedman@dbmi.columbia.edu

Journal of the American Medical Informatics Association : JAMIA
|June 10, 2004
PubMed
Summary

This study developed a natural language processing (NLP) method to automatically map clinical documents to Unified Medical Language System (UMLS) codes. The NLP approach demonstrated performance comparable to or exceeding human experts in information extraction and coding.

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

Real-world evidence for comparative safety of second-line antihyperglycemic agents in older adults with type 2 diabetes.

Nature communications·2026
Same author

A neural-symbolic AI agent system for biomedical concept mapping.

NPJ digital medicine·2026
Same author

Return of genome-informed risk-assessment results for common conditions to 23,840 adults and children: An eMERGE network study.

American journal of human genetics·2026
Same author

Accelerating real-world prediction and research in Alzheimer's: The M3AD study.

Alzheimer's & dementia : the journal of the Alzheimer's Association·2026
Same author

Comparison of Cardiovascular Events in Patients Receiving Concomitant Clopidogrel and Proton Pump Inhibitors Classified by CYP2C19 Inhibitory Potency.

Arteriosclerosis, thrombosis, and vascular biology·2026
Same author

Knowledge Engineering for Medical Vocabularies Using Large Language Models.

AMIA ... Annual Symposium proceedings. AMIA Symposium·2026

Area of Science:

  • Medical Informatics
  • Computational Linguistics
  • Clinical Documentation

Background:

  • Automated coding of clinical documents is crucial for efficient data retrieval and analysis.
  • Existing methods often require manual intervention, limiting scalability and consistency.

Purpose of the Study:

  • To develop and evaluate a natural language processing (NLP) method for automatic mapping of clinical documents to Unified Medical Language System (UMLS) codes with modifiers.
  • To quantitatively assess the performance of the NLP method against human experts.

Main Methods:

  • Adaptation of the MedLEE NLP system to generate codes from structured output (findings and modifiers).
  • Evaluation of recall and precision using two distinct test sets of 150 sentences each.
  • Comparison of NLP-generated codes against a reference standard established by multiple expert coders.

Related Experiment Videos

Main Results:

  • The NLP system achieved a recall of 0.77 for all terms and 0.83 for terms with corresponding UMLS codes.
  • System recall for extracting all terms was 0.84, comparable to expert recall (0.69-0.91).
  • The system's precision was 0.89, also competitive with expert precision (0.61-0.91).

Conclusions:

  • An NLP-based method effectively extracts clinical information and performs UMLS coding.
  • The developed method shows performance comparable to or better than human experts.
  • This automated approach enhances the suitability of coded output for effective information retrieval.