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 Video

Updated: May 12, 2026

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

Published on: August 16, 2020

Comparative Analysis of Large Language Model Performance in Appropriate Diagnostic Imaging Modality Selection.

James R Rybczyk¹, Kanhai S Amin², Varun Chamarty³

¹Renaissance School of Medicine at Stony Brook University, Stony Brook, New York.

Journal of the American College of Radiology : JACR

|May 10, 2026

Summary

Related Concept Videos

Imaging Studies II: Positron Emission Tomography and Scintigraphy

Imaging Studies II: Positron Emission Tomography and Scintigraphy

Positron Emission Tomography (PET) is a medical imaging technique that provides crucial insights into the body's physiological functions at a molecular level. It is an indispensable resource for diagnosing, staging, and monitoring various illnesses, notably cancer, neurological disorders, and cardiovascular conditions.
Fundamental Principles of PET

You might also read

Related Articles

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

Sort by

Same author

The NIH 2025 Public Access Policy: Immediate access, unequal costs.

PLoS medicine·2026

Same author

Emergency Physicians' Electronic Health Record use and High-Risk Unscheduled Return Visits: A Case-Control Study.

The Journal of emergency medicine·2026

Same author

Current Causes of Death among Children and Adolescents in the United States.

The New England journal of medicine·2026

Same author

Community-informed design of personalized patient-facing materials to facilitate ED-initiated treatment of opioid use disorder.

PEC innovation·2026

Same author

Reframing ED Boarding Through a Human Factors Sociotechnical Systems Lens.

Journal of patient safety·2026

Same author

Sex Differences in Physician Attrition from Clinical Practice Across Specialties: A Nationwide, Longitudinal Analysis.

Journal of general internal medicine·2026

Same journal

ACR Appropriateness Criteria® Myelopathy: 2026 Update.

Journal of the American College of Radiology : JACR·2026

Same journal

ACR Appropriateness Criteria® Chronic Knee Pain: Update 2026.

Journal of the American College of Radiology : JACR·2026

Same journal

Reply.

Journal of the American College of Radiology : JACR·2026

Same journal

Radiation Sensibilities: The American College of Radiology Dose Index Registry Empowers Stakeholders in Radiation Dose Optimization.

Journal of the American College of Radiology : JACR·2026

Same journal

Supply Chain Vulnerabilities in Breast Imaging: Site- and Network-Level Strategies for a Concentrated Consumable Market.

Journal of the American College of Radiology : JACR·2026

Same journal

Prostate MRI Practices and PI-RADS Use in China's Mainland: A Nationwide Assessment and Opportunities for Standardization.

Journal of the American College of Radiology : JACR·2026

See all related articles

This summary is machine-generated.

Seven large language models (LLMs) were evaluated for diagnostic imaging recommendations. While all LLMs provided appropriate imaging guidance, citation quality varied significantly, with Google

Area of Science:

Artificial Intelligence in Medicine
Medical Informatics
Clinical Decision Support Systems

Background:

Large language models (LLMs) demonstrate potential in assisting with diagnostic imaging modality selection based on American College of Radiology (ACR) criteria.
Accurate imaging recommendations are crucial for effective patient care and resource optimization.

Purpose of the Study:

To compare the performance of seven leading LLMs in providing accurate diagnostic imaging recommendations.
To assess the accuracy, clinical reasoning, and citation quality of LLM-generated recommendations using clinical vignettes.
To identify the strengths and weaknesses of different LLMs in a simulated clinical decision-making context.

Main Methods:

Fifty clinical vignettes were developed based on ACR guidelines, incorporating variations in patient presentation.

Keywords:

ACR Appropriateness Criteria ChatGPT Claude Gemini Large language model OpenEvidence clinical decision support radiology

Related Experiment Videos

Last Updated: May 12, 2026

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

Published on: August 16, 2020

LLM outputs were evaluated by blinded reviewers on imaging appropriateness, technical specificity, clinical rationale, and citation quality using a Likert scale.

Statistical analysis, including Friedman's test and Wilcoxon signed-rank testing, was employed to determine significant differences in performance.

Main Results:

All evaluated LLMs demonstrated competence in recommending appropriate imaging modalities with sound clinical justification.
No significant differences were found in the appropriateness, technical specificity, or clinical rationale scores among the LLMs.
Citation quality was highly variable, with Google's Gemini models exhibiting significant citation hallucination (80% and 76%).

Conclusions:

While LLMs can effectively guide diagnostic imaging selection, their clinical utility is contingent on reliable citation practices.
Ensuring the accuracy and relevance of cited sources is paramount for the safe and effective implementation of LLMs in clinical settings.
Further development is needed to improve the citation validity of LLMs in medical applications.