API
< Back to all clusters
[HEALTH] · South Korea · 3 sources

started · updated

AI analysis of smartwatch ECG data may detect heart dysfunction

Preliminary research presented at the American Heart Association’s Scientific Sessions suggests that artificial intelligence software analyzing ECG data from smartwatches may help identify left ventricular systolic dysfunction (LVSD), a condition involving reduced heart pumping function.

A study involving more than 2,200 adults in South Korea found that the AI model could distinguish between individuals with and without LVSD. However, researchers noted discrepancies between devices, as 9% of participants received differing results depending on the smartwatch brand used.

In a separate group of 146 adults, researchers found that requiring positive results on at least two of three consecutive days halved false-positive results compared to acting on a single positive day. While this method reduced false positives, it also resulted in missing two individuals with the condition. The findings suggest smartwatch ECGs could serve as a tool to identify those needing further screening, such as echocardiograms, though more research is required before they can replace traditional exams.

Entities

American Heart Association