Key result
ChatGPT correctly identifies and localizes ~91% of STEMIs on 12-lead ECGs.
Why the study?
Prompt diagnosis and timely reperfusion in STEMI require rapid ECG interpretation and treatment decisions, which may be aided by large language models like ChatGPT, especially in resource-limited settings.
Does ChatGPT accurately diagnose STEMI territory and recommend appropriate reperfusion therapy in patients presenting with STEMI?
Observational (n=210)
No
Does ChatGPT accurately diagnose STEMI territory and recommend appropriate reperfusion therapy in patients presenting with STEMI?
ChatGPT demonstrated high diagnostic accuracy for STEMI localization and strong concordance with real-world reperfusion triage, highlighting its potential as an AI-driven decision-support tool in resource-limited or prehospital settings.
Supports AI-assisted STEMI ECG interpretation; leaves open prospective validation before clinical adoption.
Prompt diagnosis and timely reperfusion are essential in ST-segment elevation myocardial infarction (STEMI). Large language models, such as ChatGPT, may assist non-specialist providers by rapidly interpreting electrocardiograms (ECGs) and recommending evidence-based treatment, particularly in high-volume or resource-limited settings. To assess ChatGPT’s accuracy in localizing STEMI (anterior, inferior, lateral, or true left bundle branch block [LBBB]) on 12-lead ECGs, and guiding fibrinolysis vs. primary percutaneous coronary intervention (PCI) based on estimated travel times to a tertiary center. In this retrospective single-center study, 210 STEMI patients who underwent coronary angiography were included. For each case, anonymized ECG data and the location where the ECG was recorded were input into ChatGPT. Using GPS-based travel estimates to Konya City Hospital, ChatGPT recommended fibrinolysis if transit exceeded 120 min, otherwise PCI. Diagnostic accuracy was compared with cardiologist-confirmed STEMI localization, and agreement with real-life treatment decisions was evaluated via Cohen’s kappa. ChatGPT correctly identified 192 of 210 STEMIs (91.4% accuracy; Cohen’s kappa = 0.88, 95% CI: 0.82–0.93). Of 60 patients > 120 min away, ChatGPT advised fibrinolysis in 57 (95%), matching actual fibrinolysis in 49 (82%). For 150 patients ≤ 120 min away, ChatGPT recommended PCI in 148 (98.7%), aligning with real-life management in 144 (96%). Overall concordance with real treatment decisions was 90%. By integrating ECG data with GPS-based transit times, ChatGPT achieved high accuracy in STEMI localization and guideline-concordant reperfusion recommendations. This approach may help expedite care where cardiology expertise is not immediately available.
No takes yet. Share an insight, caveat, or question.
Tezcan et al. (2026) conducted an observational in ST-segment elevation myocardial infarction (STEMI) (n=210). ChatGPT-based ECG interpretation and reperfusion triage vs. Cardiologist interpretation and real-life treatment decisions was evaluated on Diagnostic accuracy for localizing STEMI (anterior, inferior, lateral, or true left bundle branch block) (Cohen's kappa 0.88, 95% CI 0.82-0.93). ChatGPT correctly identified and localized 91.4% of ST-segment elevation myocardial infarctions on 12-lead ECGs, demonstrating strong agreement with cardiologist diagnoses (Cohen's kappa 0.88).
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:
More on Artificial intelligence in cardiologyMore from BMC Medical Informatics and Decision Making