Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
August 17, 2025Journal of Intensive Care Medicine

Diagnostic Accuracy of a Large Language Model (ChatGPT-4) for Patients Admitted to a Community Hospital Medical Intensive Care Unit: A Retrospective Case Study

View Full Paper
Ask AI
Bookmark
Share

Authors

JSJassimran SinghRBR. BohraVMVaibhavi Mukhtiar

Discussion

Loading...

Member takes

Overview

Retrospective case study found ChatGPT-4's diagnostic accuracy is 85%, indicating it could complement physician diagnoses.

Key Points

  • ChatGPT-4 achieved a diagnostic accuracy of 85%, closely following physicians' 88.3% accuracy, with no significant difference.
  • The study involved a retrospective case analysis of 120 patients in a medical intensive care unit, aiming to assess real-world diagnostic performance.
  • Using a cut-and-paste method, ChatGPT-4's diagnoses were compared to those of critical care physicians via blinded reviews for accuracy confirmation.
  • Findings indicate moderate agreement between physician and ChatGPT-4 diagnoses, suggesting potential improvement when used together.

Cite This Study

Singh et al. (2025) studied this question.

synapsesocial.com/papers/68af31ddcf1dd9ea359e779bhttps://doi.org/10.1177/08850666251368270
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1Diagnostic performance of newly developed large language models for critical illness cases: A comparative study2025
  2. 2Evaluating the Potential and Accuracy of ChatGPT-3.5 and 4.0 in Medical Licensing and In-Training Examinations: Systematic Review and Meta-Analysis2025 · 20 citations
  3. 3Evaluating the Potential and Accuracy of ChatGPT-3.5 and 4.0 in Medical Licensing and In-Training Examinations: Systematic Review and Meta-Analysis (Preprint)2024
  4. 4Evaluation of Differential Diagnosis Accuracy Using a Customized GPT-Based Artificial Intelligence Model in the Emergency Department (Preprint)2025
  5. 5Evaluating Diagnostic Performance of Laypersons, Physicians, and AI-Augmented Physicians Across Clinical Complexity Levels2025