Coronary disease & ACSCohort study
Performance of Artificial Intelligence–Powered Electrocardiogram (ECG) Analysis in Suspected ST-Segment Elevation Myocardial Infarction
- retrospectively analyzed the performance of a novel AI-ECG model
- multicenter U.S. STEMI registry (2018-2022)
- consecutive patients
Authors' conclusion
In suspected STEMI, this AI-ECG model correctly identified nearly all patients with acute coronary obstruction and most of those without AMI. If prospectively validated, this approach could improve management of patients with suspected AMI.
Study context
- Population
- Patients with cardiac catheterization laboratory activation for suspected STEMI68.3% male, median age 63 years · acute myocardial infarction (AMI) with culprit 2,076 · AMI without culprit 314 · no-AMI 133
- Participants
- 2,523 patients
- Exposure
- AI-ECG model classification OMI(+) or OMI(-) against clinical cohorts based on angiographic culprit and troponin elevation
- AMI with culpritn = 2,076
- AMI without culpritn = 314
- no-AMIn = 133
Main finding
As reported in the abstract
Correctly identified as OMI(+) among AMI with culprit patients
AMI with culprit
93.8%
Observational data: associations, not effects.
Further findings
- Correctly identified as OMI(-) among no-AMI patients
- 79.7%
Sensitivity by TIMI flow
- TIMI flow 0/1
- 96.3%
- TIMI flow 2
- 93.1%
- TIMI flow 3
- 86.9%
- P < 0.001
AUCROC
- AUCROC
- 0.952 (95% CI: 0.924-0.966)
OMI(+) within AMI without culprit cohort
- takotsubo syndrome OMI(+)
- 78%
- MI with nonobstructive coronary arteries OMI(+)
- 61%
- myopericarditis OMI(+)
- 67%
Original publication
Performance of Artificial Intelligence–Powered ECG Analysis in Suspected ST-Segment Elevation Myocardial Infarction. JACC Adv. Published 20 March 2026. doi:10.1016/j.jacadv.2026.102671
Summarised from the published abstract; every number is checked against it