JMIR Publications

ABSTRACT

Background:

Continuous monitoring of photoplethysmography (PPG) with a wearable device may aid early detection of atrial fibrillation (AF).

Objective:

We developed a ring-type wearable device (CART) to detect AF using deep learning analysis of PPG signals.

Methods:

Patients with persistent AF who underwent cardioversion were recruited prospectively. We recorded PPG signals at the finger with the CART and a conventional pulse oximeter before and after cardioversion over 15 min with each instrument. Cardiologists validated the PPG rhythms with simultaneous single-lead electrocardiograms. The PPG data were transmitted to a smartphone wirelessly and analyzed with a deep learning algorithm.

Results:

From 100 study participants, the CART generated a total of 13,038 30-s-long PPG samples (5,850 for sinus rhythm and 7,188 for AF). Using the deep learning algorithm, its diagnostic accuracy, sensitivity, specificity, positive predictive value, and negative predictive value were 96.9%, 99.0%, 94.3%, 95.6%, and 98.7%, respectively. Although the diagnostic accuracy decreased with shorter sample lengths, its accuracy was maintained at 94.7% with 10-s measurements. For sinus rhythm, the specificity decreased with higher variability of peak-to-peak intervals. However, for AF, the CART maintained consistent sensitivity, regardless of variability. Pulse rates had a relatively lesser impact on sensitivity than on specificity. Its performance was comparable to that of the conventional device when using proper thresholding.

Conclusions:

A ring-type wearable device with deep learning analysis of PPG signals could accurately diagnose AF without relying on electrocardiograms. With this device, continuous monitoring for AF among the high-risk population may be promising. Clinical Trial: ClinicalTrials.gov: NCT04023188.