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 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

Open-source offline-deployable retrieval-augmented large language model for assisting pancreatic cancer staging.

Hisashi Johno1, Akitomo Amakawa2, Atsushi Komaba2

  • 1Department of Diagnostic Radiology, Faculty of Medicine, University of Yamanashi, 1110 Shimokato, Chuo, Yamanashi, Japan. johnoh@yamanashi.ac.jp.

Japanese Journal of Radiology
|July 9, 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

Utility of serum Krebs von den Lungen-6 and surfactant protein D levels for differential diagnosis of immune checkpoint inhibitor-related pneumonitis.

Immunotherapy·2026
Same author

Diagnostic performance of commercial AI systems versus participating radiologists for pulmonary nodule detection in routine clinical practice.

Japanese journal of radiology·2026
Same author

High-fat high-sucrose diet feeding exacerbates early and complete weaning-induced steatohepatitis with ER stress-related and epigenetic changes at the Cidec regulatory region.

Nutrition (Burbank, Los Angeles County, Calif.)·2026
Same author

Deep learning-based dose prediction in proton beam therapy for hepatocellular carcinoma: comparison of network architectures and loss functions.

Journal of radiation research·2026
Same author

Serum mass spectral fingerprints with machine learning for early discrimination of infectious pneumonia in the emergency department.

BMC infectious diseases·2026
Same author

Hybrid Supercharged Antibodies: A Rational Approach to Boost Immunoassay Sensitivity via Controlled Nanoparticle Adsorption.

Langmuir : the ACS journal of surfaces and colloids·2026

This study introduces an open-source, offline retrieval-augmented large language model (RA-LLM) for pancreatic cancer staging. The RA-LLM system enhances diagnostic accuracy and transparency while preventing data leakage, demonstrating comparable performance to cloud-based models.

Area of Science:

  • Radiology and Artificial Intelligence
  • Medical Informatics
  • Oncology

Background:

  • Large language models (LLMs) show promise in radiology but face challenges like data leakage and limited transparency.
  • Developing secure and reliable AI tools for medical applications is crucial.

Purpose of the Study:

  • To develop an open-source, offline-deployable retrieval-augmented LLM (RA-LLM) system for pancreatic cancer staging.
  • To address data leakage, improve accuracy, and enhance reasoning transparency using reliable external knowledge (REK).

Main Methods:

  • Utilized local LLMs (Llama-3.2 11B, Gemma-3 27B) and a cloud-based comparator (GPT-4o mini).
  • Employed the Japanese pancreatic cancer guideline as REK for retrieval-augmented generation (RAG).
  • Evaluated system performance on 100 simulated pancreatic cancer cases, comparing RAG vs. non-RAG LLMs.
Keywords:
Large language modelsOffline deploymentOpen-source softwarePancreatic cancer stagingRetrieval-augmented generation

More Related Videos

An Orthotopic Resectional Mouse Model of Pancreatic Cancer
07:17

An Orthotopic Resectional Mouse Model of Pancreatic Cancer

Published on: September 24, 2020

Related Experiment Videos

Last Updated: Jul 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

An Orthotopic Resectional Mouse Model of Pancreatic Cancer
07:17

An Orthotopic Resectional Mouse Model of Pancreatic Cancer

Published on: September 24, 2020

Main Results:

  • RAG significantly improved TNM staging accuracy across all tested LLMs (e.g., GPT-4o mini 61% to 90%, p < 0.001).
  • Gemma-3 27B demonstrated performance comparable to GPT-4o mini, with high retrieval metrics and comparable execution times.
  • Resectability classification also showed improvement with RAG, particularly for GPT-4o mini and Llama-3.2 11B.

Conclusions:

  • An offline-deployable RA-LLM system for pancreatic cancer staging was successfully developed and open-sourced.
  • RA-LLMs offer superior performance over baseline LLMs, enhancing accuracy and transparency in radiological applications.
  • The offline-capable Gemma-3 27B provides a viable alternative to cloud-based LLMs, ensuring data security and comparable efficacy.