Task-specific versus general-purpose AI models in ECG analysis: A comparative study with emergency medicine specialists.

Journal: The American journal of emergency medicine
Published Date: Jun 27, 2025

Abstract

PURPOSE: To evaluate and compare the diagnostic accuracy of three Artificial intelligence (AI) models-GPT-4o, Canva-GPT, and ECG Reader-GPT-against emergency medicine specialists (EMSs) in electrocardiogram (ECG) interpretation using a standardized and validated test set.

Authors

  • Ertugrul Altinbilek
    University of Health Sciences, Şişli Hamidiye Etfal Training and Research Hospital, Department of Emergency Medicine, Istanbul, Turkey. Electronic address: ertugrulaltinbilek@gmail.com.
  • Adem Az
    University of Health Sciences, Haseki Training and Research Hospital, Department of Emergency Medicine, Istanbul, Turkey.
  • Ozgur Sogut
    University of Health Sciences, Haseki Training and Research Hospital, Department of Emergency Medicine, Istanbul, Turkey. Electronic address: ozgur.sogut@sbu.edu.tr.
  • Yunus Dogan
    University of Health Sciences, Haseki Training and Research Hospital, Department of Emergency Medicine, Istanbul, Turkey.
  • Tarik Akdemir
    University of Health Sciences, Haseki Training and Research Hospital, Department of Emergency Medicine, Istanbul, Turkey.
  • Erdal Belen
    University of Health Sciences, Haseki Training and Research Hospital, Department of Cardiology, Istanbul, Turkey.
  • Halil Ibrahim Biter
    University of Health Sciences, Haseki Training and Research Hospital, Department of Cardiology, Istanbul, Turkey.
  • Tugay Saricicek
    University of Health Sciences, Haseki Training and Research Hospital, Department of Emergency Medicine, Istanbul, Turkey.
  • Mehmet Ozcomlekci
    University of Health Sciences, Haseki Training and Research Hospital, Department of Emergency Medicine, Istanbul, Turkey.
  • Nurbaki Kilic
    University of Health Sciences, Haseki Training and Research Hospital, Department of Emergency Medicine, Istanbul, Turkey.

Keywords

External Resources

Popular Topics
Recent Journals