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: Apr 17, 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

1.3K

Augmenting Large Language Model With Prompt Engineering and Supervised Fine-Tuning in Non-Small Cell Lung Cancer

Ruonan Jin1, Chao Ling1, Yixuan Hou2

  • 1Liangyihui Network Technology Co, Ltd, 9/F, Tower T2, Jinheshangcheng, 140 Tianlin Road, Xuhui District, Shanghai, Shanghai, 200233, China, 86 13636599645.

JMIR AI
|April 15, 2026
PubMed

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

Author Correction: Assessment of human leukocyte antigen-based neoantigen presentation to determine pan-cancer response to immunotherapy.

Nature communications·2025
Same author

Assessment of human leukocyte antigen-based neoantigen presentation to determine pan-cancer response to immunotherapy.

Nature communications·2024
Same author

Weighting tumor-specific TCR repertoires as a classifier to stratify the immunotherapy delivery in non-small cell lung cancers.

Science advances·2021
See all related articles
Summary

This study developed an AI framework using GLM-4-Air for accurate non-small cell lung cancer (NSCLC) TNM staging. The AI model significantly outperformed GPT-4o in staging accuracy and reduced critical errors, improving clinical decision-making.

Area of Science:

  • Artificial Intelligence in Oncology
  • Medical Informatics
  • Natural Language Processing for Healthcare

Background:

  • Accurate tumor node metastasis (TNM) staging is crucial for non-small cell lung cancer (NSCLC) treatment and prognosis.
  • Traditional rule-based NLP methods struggle with the complexity and inconsistencies of clinical reporting for TNM staging.

Purpose of the Study:

  • To develop and validate an AI framework for robust and efficient NSCLC TNM staging.
  • To enhance a large language model (GLM-4-Air) using prompt engineering and supervised fine-tuning (SFT) for improved staging accuracy.

Main Methods:

  • A curated dataset of 492 NSCLC medical imaging reports with physician-validated TNM annotations was used.
  • GLM-4-Air was optimized through iterative prompt engineering and parameter-efficient SFT for T and N staging tasks.
Keywords:
AIGLM-32BGPT-4oTNM stagingartificial intelligenceclinical decision supportdiagnostic standardizationgrassroots health care improvementlarge language modelsnon-small cell lung cancerprompt engineeringsupervised fine tuning

More Related Videos

Author Spotlight: Advancements in Molecular Biomarker Testing for Non-Squamous Non-Small Cell Lung Cancer
07:59

Author Spotlight: Advancements in Molecular Biomarker Testing for Non-Squamous Non-Small Cell Lung Cancer

Published on: September 8, 2023

1.9K

Related Experiment Videos

Last Updated: Apr 17, 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

1.3K
Author Spotlight: Advancements in Molecular Biomarker Testing for Non-Squamous Non-Small Cell Lung Cancer
07:59

Author Spotlight: Advancements in Molecular Biomarker Testing for Non-Squamous Non-Small Cell Lung Cancer

Published on: September 8, 2023

1.9K
  • The hybrid model was evaluated on a held-out test set and benchmarked against GPT-4o.

    • Main Results:

    • The optimized GLM-4-Air model achieved higher staging accuracies: 92% (T), 86% (N), 92% (M), and 90% overall, surpassing GPT-4o.
    • Macro-average F1-scores were superior: 0.914 (T), 0.815 (N), 0.831 (M) for GLM-4-Air vs. 0.836, 0.620, 0.698 for GPT-4o.
    • The model demonstrated practical deployability with efficient inference and a substantial reduction in severe staging errors.

    Conclusions:

    • The hybrid AI framework offers a highly accurate, clinically reliable, and cost-efficient solution for automated NSCLC TNM staging.
    • Domain-optimized smaller models show potential for enhancing diagnostic standardization in resource-aware healthcare settings.