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

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

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

Diabetes·2026

Same author

CXR-LT 2026 Challenge: Multi-Center Long-Tailed and Zero Shot Chest X-ray Classification.

ArXiv·2026

Same author

AI-driven Abdominal Aortic Calcification Extracted From Contrast-enhanced CT Is Predictive of All-cause Mortality and Cardiovascular Events in a Large Adult Population.

Journal of computer assisted tomography·2026

Same author

Automated Delineation of Couinaud Segments at CT for Future Liver Remnant Volumetry.

Radiology. Artificial intelligence·2026

Same author

CT-based Opportunistic Screening for Adding Clinical Value: How I Do It.

Radiology·2026

Same author

Improving non-expert performance in musculoskeletal MRI protocoling through a large language model.

Scientific reports·2026

Search research articles

Related Experiment Video

Updated: Jun 7, 2025

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

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

Published on: December 6, 2024

Enhancing chest X-ray datasets with privacy-preserving large language models and multi-type annotations: A

Ricardo Bigolin Lanfredi¹, Pritam Mukherjee¹, Ronald M Summers¹

¹Imaging Biomarkers and Computer-Aided Diagnosis Laboratory, Department of Radiology and Imaging Sciences, National Institutes of Health Clinical Center, Bldg 10, Room 1C224D, 10 Center Dr, Bethesda, MD 20892-1182, USA.

Medical Image Analysis

|November 15, 2024

Summary

MAPLEZ, a new Large Language Model (LLM) system, enhances chest X-ray (CXR) report labeling by extracting detailed findings beyond simple presence. This improves the quality and utility of CXR data for research and AI development.

Keywords:

Annotation Chest x-ray Classification Large language models Medical reports Privacy-preserving

More Related Videos

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

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

Published on: April 12, 2024

Author Spotlight: A 3D Digital Model for the Diagnosis and Treatment of Pulmonary Nodules

Author Spotlight: A 3D Digital Model for the Diagnosis and Treatment of Pulmonary Nodules

Published on: May 19, 2023

Related Experiment Videos

Last Updated: Jun 7, 2025

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

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

Published on: December 6, 2024

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

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

Published on: April 12, 2024

Author Spotlight: A 3D Digital Model for the Diagnosis and Treatment of Pulmonary Nodules

Author Spotlight: A 3D Digital Model for the Diagnosis and Treatment of Pulmonary Nodules

Published on: May 19, 2023

Area of Science:

Medical Imaging Analysis
Artificial Intelligence in Healthcare
Natural Language Processing for Radiology

Background:

Current chest X-ray (CXR) report labeling methods, including rule-based systems and supervised deep learning, have limitations in label quality and adaptability.
Existing labelers often provide only binary presence labels, restricting their usefulness for advanced analysis and dataset creation.

Purpose of the Study:

To introduce MAPLEZ (Medical report Annotations with Privacy-preserving Large language model using Expeditious Zero shot answers), a novel approach for extracting and enhancing findings labels from CXR reports.
To demonstrate MAPLEZ's capability to extract not only presence/absence but also location, severity, and radiologist uncertainty for findings.

Main Methods:

Leveraging a locally executable Large Language Model (LLM) for zero-shot extraction of detailed annotations from CXR reports.
Evaluating MAPLEZ's performance on eight abnormalities across five test sets, comparing its annotation quality against existing methods.

Main Results:

MAPLEZ achieved a 3.6 percentage point (pp) increase in macro F1 score for categorical presence annotations and over 20 pp increase in F1 score for location annotations compared to competing labelers.
Utilizing MAPLEZ's enhanced and multi-type annotations improved proof-of-concept classification quality on limited-resolution CXRs, showing a 1.1 pp increase in AUROC.

Conclusions:

MAPLEZ significantly enhances the quality and detail of CXR report annotations, overcoming limitations of traditional methods.
The improved annotations generated by MAPLEZ contribute to substantial advancements in downstream AI tasks, such as image classification, particularly for datasets with limited resolution.