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

You might also read

Related Articles

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

Sort by
Same author

HeLP-BAG score: a novel data-driven scoring system for predicting post-operative 30-day mortality using 24-hour preoperative data.

BMC anesthesiology·2026
Same author

A deep-learning model for detecting choroidal metastases and predicting primary tumors from ultra-widefield fundus imaging.

Graefe's archive for clinical and experimental ophthalmology = Albrecht von Graefes Archiv fur klinische und experimentelle Ophthalmologie·2025
Same author

Current Perspectives on the Artificial Intelligence in Critical Care Medicine.

Anesthesiology clinics·2025
Same author

The South Korean health-care system in crisis.

Lancet (London, England)·2024
Same author

Early Physiologic Numerical and Waveform Characteristics of Simulated Hemorrhagic Events With Healthy Volunteers Donating Blood.

Critical care explorations·2024
Same author

Current challenges in adopting machine learning to critical care and emergency medicine.

Clinical and experimental emergency medicine·2023

Related Experiment Video

Updated: Jan 10, 2026

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

1000

Utility of Multimodal Large Language Models in Analyzing Chest X-Rays with Incomplete Contextual Information.

Choonghan Kim1, Seonhee Cho1, Joo Heung Yoon2

  • 1Kim Jaechul Graduate School of AI, Korea Advanced Institute of Science & Technology (KAIST), Daejeon, Korea.

Healthcare Informatics Research
|November 20, 2025
PubMed
Summary
This summary is machine-generated.

Multimodal large language models (LLMs) improve chest radiography report accuracy with incomplete data. Integrating text and images enhances LLM reliability for clinical decision support.

Keywords:
Clinical Decision Support SystemsInformation IntegrationLarge Language ModelsRadiographySemantics

More Related Videos

Multi-modal Pulmonary Imaging: Using Complementary Information from CT and Hyperpolarized 129Xe MRI to Evaluate Lung Structure-Function
02:09

Multi-modal Pulmonary Imaging: Using Complementary Information from CT and Hyperpolarized 129Xe MRI to Evaluate Lung Structure-Function

Published on: April 12, 2024

973
Machine Learning-Based Cough Tone Classification: Diagnostic Exploration of Chronic Obstructive Pulmonary Disease and Respiratory Tract Infections
06:22

Machine Learning-Based Cough Tone Classification: Diagnostic Exploration of Chronic Obstructive Pulmonary Disease and Respiratory Tract Infections

Published on: September 19, 2025

407

Related Experiment Videos

Last Updated: Jan 10, 2026

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

1000
Multi-modal Pulmonary Imaging: Using Complementary Information from CT and Hyperpolarized 129Xe MRI to Evaluate Lung Structure-Function
02:09

Multi-modal Pulmonary Imaging: Using Complementary Information from CT and Hyperpolarized 129Xe MRI to Evaluate Lung Structure-Function

Published on: April 12, 2024

973
Machine Learning-Based Cough Tone Classification: Diagnostic Exploration of Chronic Obstructive Pulmonary Disease and Respiratory Tract Infections
06:22

Machine Learning-Based Cough Tone Classification: Diagnostic Exploration of Chronic Obstructive Pulmonary Disease and Respiratory Tract Infections

Published on: September 19, 2025

407

Area of Science:

  • Artificial Intelligence in Radiology
  • Medical Imaging Analysis
  • Clinical Decision Support Systems

Background:

  • Large language models (LLMs) are increasingly used in clinical settings.
  • LLM performance can degrade with incomplete radiology reports.
  • Multimodal LLMs offer potential to integrate text and image data.

Purpose of the Study:

  • To evaluate multimodal LLMs for enhanced accuracy and interpretability in chest radiography reports.
  • To assess LLM robustness with incomplete data.
  • To determine if multimodal input mitigates performance loss.

Main Methods:

  • Analysis of 300 chest radiography image-report pairs from MIMIC-CXR.
  • Testing three LLMs (OpenFlamingo, MedFlamingo, IDEFICS) in text-only and multimodal formats.
  • Systematic removal of text data (20%, 50%, 80%) and evaluation with image integration.

Main Results:

  • Text-only LLMs showed performance decline with incomplete data.
  • Multimodal input significantly improved MedFlamingo and IDEFICS performance (p < 0.001).
  • Multimodal LLMs matched or exceeded text-only performance, even with missing data, and reduced hallucination rates.

Conclusions:

  • LLMs can produce suboptimal outputs with incomplete radiology data.
  • Multimodal LLMs enhance reliability and interpretability of chest radiography reports.
  • Integrating text and images strengthens LLM utility for clinical decision-making support.