AppleInsider reports today that according to the ongoing DeepHeart study from the University of California, San Francisco, and digital health startup Cardiogram, the Apple Watch heart sensor is able to achieve 97% accuracy when it comes to detecting abnormal heart rhythms.
The study uses the DeepHeart neural network to record heart rate data using wearables such as Apple Watch and Android Wear, with the aim of detecting atrial fibrillation, a form of heart arrhythmia that can cause strokes.
With both online and in-hospital components, the study uses data from 9,750 users of the Cardiogram app DeepHeart and turns the raw sensor measurements provided into fibrillation risk scores.
The end result is able to achieve 97% accuracy when compared with a traditional 12-lead ECG. This figure was already presented last year at the Heart Rhythm Society’s Heart Rhythm 2017 Conference, and has now been confirmed via peer review, with the results being published in JAMA Cardiology today.
Source: Apple Watch more effective at detecting heart condition than KardiaBand accessory, study finds