Comparison of Emergency Medicine Specialist, Cardiologist, and Chat-GPT in Electrocardiography Assessment

Comparison of Emergency Medicine Specialist, Cardiologist, and Chat-GPT in Electrocardiography Assessment

Authors: Serkan Günay, Ahmet Öztürk, Hakan Özerol, Yavuz Yiğit, Ali Kemal Erenler

Published in: American Journal of Emergency Medicine, 2024 March 16

Conclusion:

The study demonstrates GPT-4's advanced capability in ECG interpretation, highlighting its potential as a supportive tool in medical diagnostics. While it excels across everyday scenarios and presents competitive accuracy in complex cases compared to cardiologists, the integration of such AI technology could enhance diagnostic processes in emergency medicine and cardiology.

Methods:

  • Ethical approval obtained from Hitit University Medical Faculty Clinical Research Ethics Committee.
  • Utilized 40 ECG cases from "150 ECG Cases" book, divided into 20 everyday and 20 challenging questions.
  • Participants included 12 emergency medicine specialists, 12 cardiology specialists, and GPT-4, evaluated over 12 sessions.

Results:

  • GPT-4 outperformed both emergency medicine specialists and cardiologists in everyday ECG questions.
  • In challenging ECG questions, GPT-4 surpassed emergency medicine specialists but showed comparable performance to cardiologists.
  • Overall, GPT-4 demonstrated superior diagnostic accuracy in total ECG questions compared to both professional groups.
Günay, S., Öztürk, A., Özerol, H., Yiğit, Y. and Erenler, A.K., 2024. Comparison of emergency medicine specialist, cardiologist, and chat-GPT in electrocardiography assessment. The American Journal of Emergency Medicine.
Back to blog