AI Applied to Apple Watch ECG Can ID Ejection Fraction ≤40 Percent

In HealthDay News
by Healthday

Artificial intelligence algorithm applied to watch single-lead ECG identified 13 of 16 patients with left ventricular dysfunction

FRIDAY, May 13, 2022 (HealthDay News) — Single-lead electrocardiograms (ECGs) from an Apple watch interpreted by an artificial intelligence (AI) algorithm can identify ejection fraction ≤40 percent, according to a study presented at the annual meeting of the Heart Rhythm Society, held from April 29 to May 1 in San Francisco.

Itzhak Zachi Attia, Ph.D., from the Mayo Clinic in Rochester, Minnesota, and colleagues enrolled and followed subjects to examine the use of a single-lead Apple watch ECG for screening and monitoring individuals. A total of 2,454 individuals downloaded the study app. Watch ECGs acquired within one month of a clinically ordered echocardiogram were analyzed for the presence of ejection fraction ≤40 percent by AI using a single-lead model.

Subjects were enrolled from 46 states and 11 countries. The researchers found that 92 percent of participants used the app more than once (average, 2.1/month). Subjects sent 125,610 ECGs between August 2021 and February 2022. Overall, 421 subjects had at least one sinus rhythm ECG within 30 days of an echocardiogram. Of these, 16 (3.8 percent) had an ejection fraction ≤40 percent. Thirteen of the 16 patients with low ejection fraction were identified by the watch AI ECG, which had an area under the receiver operating characteristic curve of 0.875, and sensitivity and specificity of 81.2 and 81.3 percent, respectively.

“While our data are early, the test had an area under the curve of 0.88, meaning it is as good as or slightly better than a medical treadmill test,” Attia said in a statement. “AI analysis of the watch ECG is a powerful test to identify a weak heart pump.”

Press Release

More Information

Copyright © 2022 HealthDay. All rights reserved.