AI reading of smartwatch ECGs may flag reduced heart pumping function, with results varying by watch brand
A late-breaking AHA 2026 abstract tested a Medical AI Co. model for left ventricular systolic dysfunction (ejection fraction of 40% or less); requiring repeat positives cut false positives but missed some cases.
Why it matters
Left ventricular systolic dysfunction can be present without symptoms and, untreated, can lead to heart failure, so a cheap screen from a consumer watch could route more people to an echocardiogram. Study author Hak Seung Lee, chief medical officer of Medical AI Co., which developed the model, said a single smartwatch ECG is a snapshot and repeated recordings can give different results. The abstract is preliminary and not peer-reviewed; the authors say larger studies are needed to decide how results should guide further evaluation.
Key facts
- LVSD defined as ejection fraction of 40% or less
- Model developed by Medical AI Co.
- Results varied between smartwatch brands
- Late-Breaking Science Abstract 27869