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 Experiment Video

Updated: Jul 15, 2026

Lesion Explorer: A Video-guided, Standardized Protocol for Accurate and Reliable MRI-derived Volumetrics in Alzheimer's Disease and Normal Elderly
12:50

Lesion Explorer: A Video-guided, Standardized Protocol for Accurate and Reliable MRI-derived Volumetrics in Alzheimer's Disease and Normal Elderly

Published on: April 14, 2014

Data-efficient Neuroradiology MRI Imaging Protocol Prediction Using Open-weights Large Language Models.

Marius Vach1, Christian Boschenriedter2, Daniel Weiss3

  • 1Department of Diagnostic and Interventional Radiology, Medical Faculty and University Hospital Düsseldorf, Heinrich-Heine-University Düsseldorf, Düsseldorf, Germany. marius.vach@med.uni-duesseldorf.de.

Clinical Neuroradiology
|July 14, 2026
PubMed
Summary

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

Brain age gap in multiple sclerosis: associated with disability but independent of serum biomarkers.

Therapeutic advances in neurological disorders·2026
Same author

Diagnostic Potential of Apparent Diffusion Coefficient-Based Lymph Node Classification in Breast Cancer Patients Undergoing [<sup>18</sup>F]FDG-PET/MRI.

Diagnostics (Basel, Switzerland)·2026
Same author

Age, emotional burden and deep brain stimulation electrode location shape Parkinson's disease quality of life.

NPJ digital medicine·2026
Same author

Cognitive and Emotional Determinants of Subthalamic Oscillations During Freezing While Turning in Parkinson's Disease.

The European journal of neuroscience·2026
Same author

Toward Free-Breathing Renal CEST MRI: Preclinical Evaluation of a Novel Prospective Respiratory Triggering Approach.

Magnetic resonance in medicine·2026
Same author

Blood pressure management in patients receiving rescue stenting after failed endovascular treatment in large vessel occlusion acute ischaemic stroke: a multicentre registry.

European stroke journal·2026

A locally deployable large language model (LLM) can automate MRI protocol selection, enhancing data privacy and reducing annotation needs. This AI system shows feasibility for clinical workflows with minimal labeled data.

Area of Science:

  • Medical Imaging AI
  • Natural Language Processing in Healthcare
  • Radiology Workflow Optimization

Background:

  • Automated MRI protocol selection is crucial for efficient radiology workflows.
  • Current systems face challenges with data privacy, annotation burden, and scalability.
  • Large Language Models (LLMs) offer potential solutions but require careful implementation.

Purpose of the Study:

  • To assess the feasibility of a locally deployable LLM for automated MRI protocol selection.
  • To address data privacy, annotation burden, and scalability limitations in MRI protocol selection.
  • To evaluate an SOP-grounded AI system for predicting MRI protocols from order entries.

Main Methods:

  • Retrospective analysis of 598 German-language MRI order entries across brain, head/neck, and spine domains.
Keywords:
Artificial IntelligenceLarge Language ModelsMRI protocolingMagnetic resonance imaging (MRI)NeuroradiologyPrompt Optimization

More Related Videos

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

Related Experiment Videos

Last Updated: Jul 15, 2026

Lesion Explorer: A Video-guided, Standardized Protocol for Accurate and Reliable MRI-derived Volumetrics in Alzheimer's Disease and Normal Elderly
12:50

Lesion Explorer: A Video-guided, Standardized Protocol for Accurate and Reliable MRI-derived Volumetrics in Alzheimer's Disease and Normal Elderly

Published on: April 14, 2014

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

  • Development of an SOP-grounded AI system using MedGemma 27B, optimized with SIMBA.
  • Comparison of a flat LLM approach against a hierarchical system, evaluating data efficiency with varying training subsets.

    • Main Results:

    • The flat zero-shot LLM achieved 73.07% accuracy without optimization.
    • Hierarchical models with prompt optimization showed marginal improvements but did not surpass the flat approach.
    • Performance remained stable across expanded protocol classes, demonstrating robustness.

    Conclusions:

    • A locally deployable, open-weight LLM can effectively support MRI protocol selection, ensuring data privacy and requiring minimal labeled data.
    • Hierarchical routing and prompt optimization did not significantly enhance overall performance in this dataset.
    • Findings support further investigation in human-in-the-loop clinical settings.