Artificial Intelligence in detection of acute coronary occlusion in NSTEMI patients

The prognosis of patients with MI has improved significantly with the recognition that early reperfusion is critical, particularly since timely percutaneous coronary intervention (PCI) became widely adopted. The invasive reperfusion era also reshaped MI diagnostics, shifting the paradigm from Q-wave vs. Non-Q-wave MI to ST-Elevation Myocardial Infarction (STEMI) vs. Non-ST-Elevation Myocardial Infarction (NSTEMI). The current ST-segment elevation myocardial infarction (STEMI) and non-ST-segment elevation myocardial infarction (NSTEMI) paradigm have long been the cornerstone of myocardial infarction (MI) care but fail to identify many patients with acute coronary occlusion (ACO), delaying treatment and worsening outcomes. This limitation is increasingly important, since NSTEMI now represents the majority of presentations accounting for roughly 70% of AMI worldwide and many of these occlusive events are managed with delays contributing to worse outcomes. Adding to this challenge, substantial inter-physician variability in ECG interpretation for ACO has been demonstrated. In this review, we highlight recent advances using the artificial intelligence in the evaluation of patients presenting with ECG changes suggestive of NSTEMI and evaluate its role in the detection of NSTEMI patients with acute coronary occlusion.