Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Videos

Auditing concept categorizations in the UMLS.

Huanying Gu¹, Yehoshua Perl, Gai Elhanan

¹Department of Health Informatics, University of Medicine and Dentistry of NJ, Newark, NJ 07107, USA. guhy@umdnj.edu

Artificial Intelligence in Medicine

|June 9, 2004

Summary

This summary is machine-generated.

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

Digital health and consumer health informatics: past and future.

Medical research archives·2026

Same author

The association of prenatal adiposity characteristics with early childhood overweight and obesity: findings from a large and diverse mother-child cohort.

International journal of obesity (2005)·2026

Same author

An Interoperable Vaccine Record: A Roadmap to Realization.

Vaccines·2026

Same author

Which decisions affect cohort distribution in COVID-19 data analytics?

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

Same author

Benchmarking large language models on the United States medical licensing examination for clinical reasoning and medical licensing scenarios.

Scientific reports·2025

Same author

Advancing the science of genomic learning healthcare systems.

Learning health systems·2025

Same journal

Real-time EEG-based epileptic seizure prediction using artificial intelligence: A systematic review.

Artificial intelligence in medicine·2026

Same journal

R-peak detection and ECG data compression scheme based on empirical mode decomposition and wavelet transform.

Artificial intelligence in medicine·2026

Same journal

CastNet: A three-channel EEG-based deep learning model for cross-subject depression detection.

Artificial intelligence in medicine·2026

Same journal

State-of-the-art TinyML approaches for colorectal cancer detection: Current advances, challenges, and future directions.

Artificial intelligence in medicine·2026

Same journal

JRadiEvo: A Japanese radiology report generation model enhanced by evolutionary optimization of model merging.

Artificial intelligence in medicine·2026

Same journal

Causally-informed deep learning towards explainable and generalizable outcome prediction in critical care.

Artificial intelligence in medicine·2026

See all related articles

This study introduces a novel auditing technique to detect errors and inconsistencies in the Unified Medical Language System (UMLS) semantic network. The method efficiently identifies miscategorized biomedical concepts, improving data accuracy.

Area of Science:

Biomedical Informatics
Knowledge Representation
Data Quality Management

Background:

The Unified Medical Language System (UMLS) integrates extensive biomedical terminologies, assigning semantic types to concepts.
Integration processes can introduce categorization errors and inconsistencies within the UMLS Semantic Network.
Ensuring the accuracy of semantic typing is crucial for reliable biomedical data integration and analysis.

Purpose of the Study:

To develop and present an effective auditing technique for identifying errors and inconsistencies in UMLS semantic type categorization.
To improve the quality and reliability of the UMLS Semantic Network through systematic error detection.
To analyze the types and prevalence of errors within the UMLS semantic categorization.

Main Methods:

Related Experiment Videos

The technique involves expert review of 'pure intersections' derived from a metaschema, a simplified view of the UMLS Semantic Network.
A divide and conquer strategy is employed, treating small and large intersections differently to optimize review efficiency.
Concepts within small pure intersections (1-10 concepts) were exhaustively reviewed, while larger intersections were screened for semantic soundness.

Main Results:

A significant number of categorization errors and inconsistencies were identified through the review of 657 small pure intersections.
The analysis revealed various types of errors, with detailed findings presented in the paper.
Semantically suspicious larger pure intersections were flagged, and their concepts were subsequently reviewed.

Conclusions:

The proposed auditing technique is effective in detecting errors and inconsistencies within the UMLS Semantic Network.
The findings highlight the ongoing need for robust quality assurance in large-scale biomedical knowledge integration.
This work contributes to enhancing the accuracy and utility of the UMLS for biomedical research.