View on Pitt Calendar

Topic Overview:

Prehospital detection and risk stratification of acute coronary syndrome remains a longstanding challenge in clinical practice. The 12-lead electrocardiogram (ECG) is readily available during initial patient evaluation, but current rule-based interpretation systems lack sufficient accuracy. Using a first-of-its-kind prehospital ECG database of patients calling 9-1-1 for chest pain (n = 4,132), Al-Zaiti and colleagues evaluated the performance of engineered, novel temporal-spatial ECG features and machine-learning techniques for the prediction of underlying acute myocardial ischemia. Their models outperformed commercial interpretation software and experienced clinicians in diagnosing acute coronary syndrome; detecting and localizing actionable coronary culprit lesions; and predicting 30-day major adverse cardiac events. Al-Zaiti and colleagues’ current efforts focus on building these AI-enhanced diagnostics into a prototype intelligent-decision support system for subsequent prospective testing. An ultra-early, ECG-based clinical decision support tool, when combined with the judgment of trained emergency personnel, would help to improve clinical outcomes and reduce unnecessary costs in patients with chest pain.