New ‘Queen of Hearts’ AI aims to detect hidden-blockage heart attacks

Katie covers the impact of health technology on patients, clinicians, and businesses. Her stories explore the price tag of clinical AI, digital health at the FDA, and the boom in direct-to-consumer telehealth. Confidential tips can be sent on Signal at palmer.01.

A flurry of artificial intelligence algorithms have promised to help identify heart conditions early. By analyzing an electrocardiogram, a common 10-second test of the heart’s activity, AI has been cleared to screen for common arrhythmias and structural heart conditions, and algorithms in development aim to pinpoint patients at highest risk of sudden cardiac arrest. 

While many of those models are focused on early diagnosis and prevention, AI also might help a patient when they’re having a heart attack. Last week, the Food and Drug Administration issued a rare de novo classification to a model from Powerful Medical that aims to help triage patients to the right care when they’re having chest pain. The vast majority of AI devices go to market through the FDA’s 510(k) clearance pathway; fewer than 10 each year receive de novo classification, which typically requires more clinical evidence. 

The most severe kind of heart attack occurs when a coronary artery, which supplies blood to the heart, is completely blocked off. Those patients need to get to a cardiac catheterization lab as soon as possible to open up the blockage and restore oxygen to the heart. On an EKG, doctors can identify these kinds of myocardial infarctions from their characteristic “ST-segment elevations.”