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 Concept Videos

Imaging Studies I: CT and MRI01:14

Imaging Studies I: CT and MRI

455
Introduction: MRI and CT scans are crucial advancements in medical imaging techniques, playing a vital role in diagnosing conditions related to the gastrointestinal (GI) system. Each scan serves distinct purposes, targets specific areas, and requires unique nursing duties.
Description of the Procedures
Computed Tomography (CT) scan:
Computed Tomography (CT) scans use X-ray technology to generate detailed images of bones, organs, and tissues. During the scan, the patient lies on a moving table...
455
Imaging Studies III: Computed Tomography01:27

Imaging Studies III: Computed Tomography

56
DefinitionComputed Tomography (CT) of the genitourinary (GU) tract is a non-invasive imaging modality that utilizes X-rays and computer processing to generate detailed cross-sectional images of the urinary system, encompassing the kidneys, ureters, bladder, and adjacent structures such as the adrenal glands.PurposeCT scans of the GU tract serve several diagnostic and therapeutic purposes, including:Diagnosis of Urinary Tract Diseases: Detects kidney stones, tumors, cysts, and congenital...
56
Imaging Studies II: Positron Emission Tomography and Scintigraphy01:25

Imaging Studies II: Positron Emission Tomography and Scintigraphy

239
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
239
Imaging Studies for Cardiovascular System V: CT01:28

Imaging Studies for Cardiovascular System V: CT

71
Cardiac computed tomography (CT) scanning is an advanced cardiac imaging technique that utilizes CT technology, with or without intravenous (IV) contrast, to produce accurate cross-sectional virtual slices of specific areas of the heart, coronary circulation, and major blood vessels such as the aorta, pulmonary veins, and arteries. The computer processes these slices to generate three-dimensional images. Multidetector CT (MDCT) is a rapid form of CT scanning that captures multiple slices...
71

You might also read

Related Articles

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

Sort by
Same author

Phage-guided rapid enrichment of copper-protocatechuic acid nanoparticles for targeted antibacterial therapy.

Journal of nanobiotechnology·2026
Same author

Utility of Pancreatic Perivascular Adipose Tissue as a CT Imaging Biomarker for Diagnosing Type 2 Diabetes.

Diabetes·2026
Same author

Knowledge-guided contextual gene set analysis with large language models.

Bioinformatics (Oxford, England)·2026
Same author

MedHopQA: A Disease-Centered Multi-Hop Reasoning Benchmark and Evaluation Framework for LLM-Based Biomedical Question Answering.

ArXiv·2026
Same author

Toward Multimodal Conversational AI for Age-Related Macular Degeneration.

ArXiv·2026
Same author

Versatile hollow Ca<sup>2+</sup>-phenolic nanoparticles for intracellular delivery of diverse bioactive molecules and CRISPR-Cas9 genome editing.

Biomaterials·2026

Related Experiment Video

Updated: Sep 17, 2025

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness
03:14

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness

Published on: December 6, 2024

692

How well do multimodal LLMs interpret CT scans? An auto-evaluation framework for analyses.

Qingqing Zhu1, Benjamin Hou1, Tejas Sudarshan Mathai2

  • 1National Library of Medicine (NLM), National Institutes of Health (NIH), 8600 Rockville Pike, Bethesda, 20894, MD, USA.

Journal of Biomedical Informatics
|June 27, 2025
PubMed
Summary
This summary is machine-generated.

A new framework, GPTRadScore, uses GPT-4 to evaluate multimodal large language models (MLLMs) for CT imaging analysis. This method correlates well with clinician assessments and shows fine-tuning improves model accuracy in medical diagnostics.

Keywords:
CT scansEvaluationFine-tuningLanguage modelsMedical imagingMultimodal

More Related Videos

Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images
14:08

Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images

Published on: April 13, 2013

42.8K
Author Spotlight: An Efficient and Robust Software for Automated Fusion of Multiple Preclinical Imaging Modalities
07:13

Author Spotlight: An Efficient and Robust Software for Automated Fusion of Multiple Preclinical Imaging Modalities

Published on: October 27, 2023

1.4K

Related Experiment Videos

Last Updated: Sep 17, 2025

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness
03:14

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness

Published on: December 6, 2024

692
Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images
14:08

Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images

Published on: April 13, 2013

42.8K
Author Spotlight: An Efficient and Robust Software for Automated Fusion of Multiple Preclinical Imaging Modalities
07:13

Author Spotlight: An Efficient and Robust Software for Automated Fusion of Multiple Preclinical Imaging Modalities

Published on: October 27, 2023

1.4K

Area of Science:

  • Artificial Intelligence in Medical Imaging
  • Radiology and Medical Diagnostics
  • Natural Language Processing in Healthcare

Background:

  • Multimodal large language models (MLLMs) show promise for interpreting medical images.
  • Existing evaluation metrics for MLLMs in radiology lack clinical accuracy.
  • There is a need for a robust framework to assess MLLM performance in generating clinically relevant findings from CT scans.

Purpose of the Study:

  • To introduce GPTRadScore, a novel evaluation framework utilizing LLMs for assessing MLLM performance in CT imaging.
  • To compare GPTRadScore against traditional metrics for evaluating clinical accuracy.
  • To evaluate the performance of leading MLLMs (GPT-4V, Gemini Pro Vision, LLaVA-Med, RadFM) in interpreting CT findings.

Main Methods:

  • A retrospective study using a subset of the DeepLesion dataset to evaluate MLLMs.
  • Development of GPTRadScore leveraging GPT-4 to assess generated descriptions (location, body part, type).
  • Fine-tuning of RadFM on a specialized DeepLesion subset to improve complex finding interpretation and reassessment using GPTRadScore.

Main Results:

  • GPTRadScore demonstrated high correlation with clinician assessments (Pearson's r: 0.75-0.91), outperforming traditional metrics (BLEU, METEOR, ROUGE).
  • GPT-4V and Gemini Pro Vision showed superior performance among evaluated MLLMs, though dataset limitations persist.
  • Fine-tuning RadFM significantly improved accuracy: location (3.41% to 12.8%), body part (29.12% to 53%), and type (9.24% to 30%).

Conclusions:

  • GPT-4 serves as a reliable metric for evaluating MLLMs in radiological diagnostics, correlating well with expert assessments.
  • Fine-tuning approaches are effective in enhancing the descriptive accuracy of LLM-generated medical imaging findings.
  • GPTRadScore offers a clinically informed and reliable method for assessing MLLMs in medical imaging analysis.