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 Videos

Large Language Model Versus Multidisciplinary Team: Feasibility Study of Pancreatic Cancer Management

Zhuoran Liu1, Xin Zhao2, Tianyang Mao1

  • 1Department of Hepato-Pancreato-biliary Surgery, People's Hospital of Leshan, No.238 Huian street, Shizhong District, Leshan, Sichuan Province, 614000, China, 86 833 211 9306.

Journal of Medical Internet Research
|June 30, 2026
PubMed
Summary

Related Concept Videos

Chronic Pancreatitis II: Collaborative Care01:29

Chronic Pancreatitis II: Collaborative Care

The management of chronic pancreatitis is multifaceted, involving a comprehensive approach that includes thorough assessment, diagnostic testing, and a variety of management strategies.
Assessment:

You might also read

Related Articles

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

Sort by
Same author

The impact of coronavirus disease 2019 (COVID-19) on liver injury in China: A systematic review and meta-analysis.

Medicineยท2021
See all related articles

ChatGPT-5.2 accurately matched multidisciplinary team management categories for 95.2% of pancreatic cancer cases. While promising for exploratory use, the model requires version control and a refined endpoint for clinical application.

Area of Science:

  • Oncology
  • Artificial Intelligence
  • Medical Informatics

Background:

  • Pancreatic cancer management is complex, often requiring multidisciplinary team (MDT) input.
  • Evaluating the utility of artificial intelligence (AI) tools in supporting clinical decision-making is an active area of research.

Purpose of the Study:

  • To assess the performance of a chat interface model (ChatGPT-5.2) in categorizing pancreatic cancer management plans.
  • To determine the concordance between AI-generated management categories and expert MDT decisions.

Main Methods:

  • A retrospective analysis of 125 complete structured pancreatic cancer cases from a multidisciplinary team.
  • Comparison of the AI model's initial management category assignment against the MDT's established category.
Keywords:
LLMMDTartificial intelligenceclinical decision supportlarge language modelsmultidisciplinary teampancreatic cancer

Related Experiment Videos

Main Results:

  • The ChatGPT-5.2 model achieved a high agreement rate, matching the MDT's broad initial management category in 119 out of 125 cases (95.2%).
  • Confidence intervals for the agreement rate were 95% CI 89.8%-98.2%.

Conclusions:

  • The AI model demonstrates significant potential in aligning with expert consensus for pancreatic cancer management categories.
  • Limitations include a coarse endpoint and lack of version control, suggesting current findings support only supervised exploratory use.