Automated Detection of Atrial Fibrillation Through Smartwatch Technology
Atrial fibrillation affects 9% of US adults older than 65.
ORLANDO — Researchers have confirmed that the Kardia Band (KB), a specialized smartwatch band, and its accompanying app can accurately detect atrial fibrillation and distinguish it from sinus rhythm. The research was presented at the American College of Cardiology 67th Annual Scientific Session, held March 10-12, 2018, in Orlando, Florida.
Researchers tested the KB and its algorithm against physician-interpreted 12-lead electrocardiogram (ECG) readings, comparing their respective use in differentiating between sinus rhythm and atrial fibrillation. The study population involved 100 patients who were undergoing elective cardioversion for atrial fibrillation. For eligible pre-screened patients, the researchers then simultaneously recorded KB and ECG strips, both before and after cardioversion (n=169).
The app that accompanies the KB interprets results using an automated algorithm. To measure the accuracy and sensitivity of the KB method, the algorithm's performance was compared to physician analysis twice: first, analysis of the simultaneous ECG, and then separately, analysis of the KB recordings alone by 2 blinded electrophysiologists.
The algorithm and KB device compared fairly well with the physician-analyzed ECG and proved able to differentiate sinus rhythm from atrial fibrillation with 93% sensitivity, 84% specificity, and a K coefficient of 0.77. In 113 cases, both the algorithm and the blinded electrophysiologists analyzed the results of the KB recordings to determine sinus rhythm vs atrial fibrillation. Per the researchers, in these cases “agreement was excellent with a K coefficient of 0.88.”
Importantly, in 59 cases, the algorithm found the recordings noninterpretable, but the electrophysiologists were correctly able to diagnose sinus rhythm vs atrial fibrillation with 100% sensitivity, 80% specificity, and a K coefficient of 0.74.
Clinically, these results suggest that the KB device and app would be useful in the screening process for patients with atrial fibrillation before undergoing cardioversion. Combining the device with physician readings of noninterpretable strips could help detect sinus rhythm vs atrial fibrillation, preventing unnecessary procedures.
Bumgarner J, Lambert C, Cantillon D, et al. Assessing the accuracy of an automated atrial fibrillation detection algorithm using novel smartwatch technology among patients presenting for elective cardioversion. Presented at: American College of Cardiology 67th Annual Scientific Session & Expo; March 10-12, 2018; Orlando, FL. Abstract 12500.