Artificial Intelligence–Powered Electrocardiogram Detecting Culprit Vessel Blood Flow Abnormality: AI-ECG TIMI Study Design and Rationale

Abstract

Background: The 12-lead electrocardiogram (ECG) is the gold standard for detecting patients who will benefit from emergent revascularization due to occlusive myocardial infarction (OMI). However, the pathophysiology of acute coronary syndromes (ACS) is dynamic, and nearly half of patients with OMI do not present with typical ST elevation or have dynamic ECG changes due to spontaneous recanalization before invasive coronary angiography (ICA). Recently, an ECG-based artificial intelligence (AI) model was developed using expert interpretation of OMI. However, its performance is limited to retrospective evaluation of ECGs recorded minutes to hours before ICA. Methods: The AI-ECG thrombolysis in myocardial infarction (TIMI) study is an investigator-initiated prospective multicenter registry planning to enroll over 700 consecutive patients with ACS undergoing ICA in 9 centers across Europe. For all participants, a standard 10-second 12-lead ECG will be recorded at the time of coronary angiography. The primary end point is the AI model's ability to identify patients with an actively occluded (TIMI 0-1) culprit coronary artery at the time of invasive coronary angiography using only single-standard 12-lead ECGs. Standardized angiograms will be used as a reference standard. Conclusions: AI-ECG TIMI is the first prospective registry of consecutive patients with ACS with standard 12-lead ECGs recorded at the very moment of ICA. This study will help characterize ECG findings of abnormal myocardial perfusion due to acute active ischemia and prospectively validate an AI model's ability to detect them.

Ähnliche Arbeiten