Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
September 10, 2025International Journal of Emergency MedicineOpen Access

Comparative performance of ChatGPT, Gemini, and final-year emergency medicine clerkship students in answering multiple-choice questions: implications for the use of AI in medical education

View Full Paper
Ask AI
Bookmark
Share

Authors

SAShaikha Al-ThaniSAShahzad AnjumZBZain A. Bhutta

Discussion

Loading...

Member takes

Overview

Analysis reveals students outperform AI models ChatGPT and Gemini in MCQs, suggesting limits in AI for medical education.

Key Points

  • Final-year emergency medicine students demonstrated higher accuracy than ChatGPT and Gemini.
  • Students scored an overall accuracy of 79.4%, while ChatGPT achieved 72.5% and Gemini only 54.4%.
  • A comparative analysis of 160 MCQs examined performance on both text-only and image-based questions.
  • Significant gaps in performance emphasize the current limitations of AI in visual interpretation and clinical reasoning.

Cite This Study

Al-Thani et al. (2025) studied this question.

synapsesocial.com/papers/68c24236b210217d647a288fhttps://doi.org/10.1186/s12245-025-00949-6
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. 1Performance of Generative AI Models on Cardiology Practice in Emergency Service: A Pilot Evaluation of GPT-4.o and Gemini-1.5-Flash2025
  2. 2Evaluating and comparing student responses in examinations from the perspectives of human and artificial intelligence (GPT-4 and Gemini)2025 · 1 citations
  3. 3The performance of ChatGPT on medical image-based assessments and implications for medical education2025 · 12 citations
  4. 4Comparing AI-Generated Responses: A Study on ChatGPT, Gemini, and Copilot in Education2025 · 4 citations
  5. 5Evaluating the Potential and Accuracy of ChatGPT-3.5 and 4.0 in Medical Licensing and In-Training Examinations: Systematic Review and Meta-Analysis2025 · 22 citations