Ambulatory Atrial Fibrillation Detection and Quantification by a wrist-worn AI-device Compared to Standard Holter Monitoring

Abstract

Introduction Timely detection of atrial fibrillation (AF) is crucial for the prevention of serious consequences such as stroke and heart failure, yet it remains challenging due to its often asymptomatic or paroxysmal nature. Wearable devices with artificial intelligence algorithms offer a promising solution. Methods AF detection by the CardioWatch 287-2 (CW2), a wrist-worn photoplethysmography (PPG) and single-lead ECG device, was compared to 24-hour Holter recordings. Patient compliance, AF prevalence and AF burden were evaluated for a further 27 days. Results Data from 150 participants (mean age 64 ±12 SD; 41% female) were analysed. The CW2’s PPG and single-lead ECG algorithms achieved a specificity above 98% and sensitivity above 95% for AF detection, and a correlation of 99% for AF burden, compared to 24-hour Holter. AF prevalence increased from 14.7% (24-hour Holter) to 26.7% (28-day CW2). Conclusion The wrist-worn CW2 device showed promising performance in detecting AF and determining AF burden.