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 Concept Videos

You might also read

Related Articles

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

Sort by
Same author

Beyond accuracy metrics: Toward responsible integration of large language models in pediatric diagnostic reasoning.

Pediatric investigation·2026
Same author

The hidden myocardium: Why intraoperative hypotension should be treated as myocardial injury prevention rather than a transient anaesthetic variable.

Journal of perioperative practice·2026
Same author

The Responsible Health AI Readiness and Maturity Index (RHAMI): Applications for a Global Narrative Review of Leading AI Use Cases in Public Health Nutrition.

Nutrients·2026
Same author

Performance of DeepSeek and GPT Models on Pediatric Board Preparation Questions: Comparative Evaluation.

JMIR AI·2025
Same author

Cardiotoxicity in Pediatric Cancer Survivorship: Retrospective Cohort Study.

JMIRx med·2025
Same author

Advancing Early Detection of Major Depressive Disorder Using Multisite Functional Magnetic Resonance Imaging Data: Comparative Analysis of AI Models.

JMIRx med·2025

Related Experiment Video

Updated: May 21, 2025

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

474

Large Language Models for Pediatric Differential Diagnoses in Rural Health Care: Multicenter Retrospective Cohort

Masab Mansoor1, Andrew F Ibrahim2, David Grindem3

  • 1Edward Via College of Osteopathic Medicine, 4408 Bon Aire Dr, Monroe, LA, 71203, United States, 1 5045213500.

Jmirx Med
|March 19, 2025
PubMed
Summary

A fine-tuned GPT-3 model offers diagnostic support comparable to pediatricians in rural settings, showing promise for improving pediatric differential diagnosis. This AI tool demonstrated high accuracy, especially for common conditions, aiding rural healthcare providers.

Keywords:
GPTLLMMLNLPartificial intelligencegenerative pretrained transformerlanguage modellarge language modelmachine learningnatural language processingpediatrics

More Related Videos

Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application
05:56

Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application

Published on: April 14, 2023

2.3K
Implementation of a Real-Time Psychosis Risk Detection and Alerting System Based on Electronic Health Records using CogStack
07:31

Implementation of a Real-Time Psychosis Risk Detection and Alerting System Based on Electronic Health Records using CogStack

Published on: May 15, 2020

7.0K

Related Experiment Videos

Last Updated: May 21, 2025

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

474
Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application
05:56

Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application

Published on: April 14, 2023

2.3K
Implementation of a Real-Time Psychosis Risk Detection and Alerting System Based on Electronic Health Records using CogStack
07:31

Implementation of a Real-Time Psychosis Risk Detection and Alerting System Based on Electronic Health Records using CogStack

Published on: May 15, 2020

7.0K

Area of Science:

  • Artificial Intelligence in Healthcare
  • Pediatric Diagnostics
  • Rural Health Technology

Background:

  • Rural healthcare providers face challenges including limited specialist access and high patient volumes.
  • Accurate diagnostic support tools are crucial for rural pediatric care.
  • Large language models (LLMs) show potential for clinical decision support but need study in pediatric differential diagnosis.

Purpose of the Study:

  • To evaluate the diagnostic accuracy and reliability of a fine-tuned GPT-3 model.
  • To compare the GPT-3 model's performance against board-certified pediatricians.
  • To assess the model's utility in rural pediatric healthcare settings.

Main Methods:

  • A multicenter retrospective cohort study analyzed 500 pediatric encounters (ages 0-18) from rural Louisiana.
  • A GPT-3 model (DaVinci) was fine-tuned and tested on 150 encounters.
  • Performance was compared against diagnoses from five board-certified pediatricians using accuracy, sensitivity, and specificity.

Main Results:

  • The GPT-3 model achieved 87.3% accuracy, comparable to pediatricians' 91.3% (P=.47).
  • Sensitivity was 85% and specificity 90% for the model.
  • Performance remained consistent across age groups and common complaints, with slightly lower accuracy for rare diagnoses (80% vs. 85% for pediatricians).

Conclusions:

  • A fine-tuned GPT-3 model provides diagnostic support comparable to pediatricians in rural healthcare.
  • The AI model is particularly effective for common pediatric presentations.
  • Further validation in diverse populations is recommended before widespread clinical implementation.