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

Application of recently developed computer algorithm for automatic classification of unstructured radiology reports:

Keith J Dreyer1, Mannudeep K Kalra, Michael M Maher

  • 1Division of Computing and Information Services, Department of Radiology, Massachusetts General Hospital and Harvard Medical School, 100 Charles River Plaza, Suite 471, Cambridge St, Boston, MA 02114, USA. kdreyer@partners.org

Radiology
|December 14, 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

Evaluation of an Artificial Intelligence Model for Identification of Intracranial Hemorrhage Subtypes on Computed Tomography of the Head.

Stroke (Hoboken, N.J.)·2026
Same author

Large language models as cost-conscious decision aids in emergency medicine: protocol support for imaging in lower back pain.

Emergency radiology·2025
Same author

Evaluation of an artificial intelligence model for opportunistic Agatston scoring on non-gated chest computed tomography.

Scientific reports·2025
Same author

Regulating Generative AI in Radiology Practice: A Trilaminar Approach to Balancing Risk with Innovation.

Academic radiology·2025
Same author

Detection of Hypertrophic Cardiomyopathy on Electrocardiogram Using Artificial Intelligence.

Circulation. Heart failure·2025
Same author

PRECISE framework: Enhanced radiology reporting with GPT for improved readability, reliability, and patient-centered care.

European journal of radiology·2025

Lexicon Mediated Entropy Reduction (LEXIMER) accurately classifies unstructured radiology reports for clinically important findings and recommendations. This automated engine shows high accuracy, making it a valuable tool for radiology report analysis.

Area of Science:

  • Radiology and Medical Imaging
  • Natural Language Processing
  • Health Informatics

Background:

  • Unstructured radiology reports contain critical clinical information that is challenging to extract and analyze.
  • Automated analysis of radiology reports can improve efficiency and consistency in clinical decision-making.
  • Information theory-based algorithms offer novel approaches to analyzing complex textual data.

Purpose of the Study:

  • To validate the accuracy of Lexicon Mediated Entropy Reduction (LEXIMER), an information theory-based algorithm.
  • To assess LEXIMER's capability in classifying unstructured radiology reports based on key findings and recommendations.
  • To establish LEXIMER as a reliable tool for independent analysis of radiology report content.

Main Methods:

Related Experiment Videos

  • A dataset of 1059 de-identified radiology reports from various modalities was used.
  • Two radiologists independently categorized reports for clinically important findings (F(T)) and recommendations (R(T)).
  • LEXIMER algorithm was employed to categorize reports, and its performance was statistically assessed against radiologist consensus.

    • Main Results:

    • LEXIMER demonstrated high accuracy in classifying reports for clinically important findings (97.5%) and recommendations (99.6%).
    • Sensitivity and specificity for LEXIMER were also notably high, indicating robust performance.
    • Strong interobserver concordance (kappa = 0.9) was observed between the two radiologists.

    Conclusions:

    • LEXIMER is a highly accurate automated tool for analyzing unstructured radiology reports.
    • The algorithm effectively evaluates the presence of clinically important findings and rates of recommendations.
    • LEXIMER offers a reliable method for the automated assessment of radiology report content.