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

Positron Emission Tomography

Positron Emission Tomography

Positron emission tomography (PET) is a medical imaging technique involving radiopharmaceuticals — substances that emit short-lived radiation. Although the first PET scanner was introduced in 1961, it took 15 more years before radiopharmaceuticals were combined with the technique and revolutionized its potential.
One of the main requirements of a PET scan is a positron-emitting radioisotope, which is produced in a cyclotron and then attached to a substance used by the part of the body...

You might also read

Related Articles

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

Sort by

Same author

Large Language Models for Multidisciplinary Tumor Board Decision-Making in Primary Liver Tumors: A Retrospective Single-Center Study.

Cancers·2026

Same author

Symptom-Only Localization of Brainstem Ischemia Using Large Language Models Versus Neurologists in Diffusion-Weighted Imaging-Positive Cases: Retrospective Single-Center Study.

JMIR formative research·2026

Same author

Generalizable and explainable deep learning for brain MRI: a multi-cohort evaluation of 3D architectures for age and sex prediction.

Brain informatics·2026

Same author

Erratum for: Associations of MRI-derived Paraspinal IMAT and LMM with Cardiometabolic Risk Factors: Results from a German Cohort.

Radiology·2026

Same author

Counterfactual Diffusion Models Provide Interpretable Explanations of Artificial Intelligence Models in Pathology.

Cancer research·2026

Same author

Towards autonomous medical artificial intelligence agents.

Nature·2026

Search research articles

Related Experiment Video

Updated: Jun 10, 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

Automatic structuring of radiology reports with on-premise open-source large language models.

Piotr Woźnicki¹, Caroline Laqua², Ina Fiku²

¹Department of Diagnostic and Interventional Radiology, University Hospital Würzburg, Würzburg, Germany. woznicki_p@ukw.de.

European Radiology

|October 10, 2024

Summary

A locally hosted large language model (LLM) can automatically structure radiology reports, achieving performance comparable to human readers. This technology aids in adopting structured reporting without disrupting radiologist workflows.

Keywords:

Chest radiography Large language models Structured reporting

More Related Videos

A Metadata Extraction Approach for Clinical Case Reports to Enable Advanced Understanding of Biomedical Concepts

A Metadata Extraction Approach for Clinical Case Reports to Enable Advanced Understanding of Biomedical Concepts

Published on: September 20, 2018

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

Related Experiment Videos

Last Updated: Jun 10, 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

A Metadata Extraction Approach for Clinical Case Reports to Enable Advanced Understanding of Biomedical Concepts

A Metadata Extraction Approach for Clinical Case Reports to Enable Advanced Understanding of Biomedical Concepts

Published on: September 20, 2018

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

Area of Science:

Radiology Informatics
Natural Language Processing
Artificial Intelligence in Medicine

Background:

Structured reporting enhances radiology report comparability, readability, and detail.
Integrating structured reporting into clinical workflows remains a challenge.
Large language models (LLMs) offer potential for automated data extraction from free-text reports.

Purpose of the Study:

To evaluate an on-premise, privacy-preserving LLM for automatically structuring free-text radiology reports.
To assess the performance of a locally hosted LLM in generating structured data from narrative reports.
To compare LLM performance against human readers in structuring radiology reports.

Main Methods:

Developed a controlled approach for a locally hosted Llama-2-70B-chat model.
Utilized a retrospective dataset of 202 English (MIMIC-CXR) and 197 German chest radiograph reports.
Employed a senior radiologist-defined template with 48 question-answer pairs and Bayesian inference for performance evaluation (Matthews correlation coefficient).

Main Results:

The LLM generated valid structured reports across all cases.
Average Matthews correlation coefficient (MCC) was 0.75 for English and 0.66 for German reports.
LLM performance was comparable to human readers, with MCC differences within the region of practical equivalence (ROPE) for both languages.

Conclusions:

Open-source LLMs hosted locally can automate the structuring of free-text radiology reports with near-human accuracy.
LLM performance demonstrated variations in semantic understanding across different languages and imaging findings.
Automating report structuring can facilitate structured data capture while preserving radiologist workflow.