Validation of an algorithm for continuous monitoring of atrial fibrillation using a consumer smartwatch

Abstract

BACKGROUND Consumer devices with broad reach may be useful in screening for atrial fibrillation (AF) in appropriate populations. However, currently no consumer devices are capable of continuous monitoring for AF.OBJECTIVE The purpose of this study was to estimate the sensitivity and specificity of a smartwatch algorithm for continuous detection of AF from sinus rhythm in a free-living setting.METHODS We studied a commercially available smartwatch with photoplethysmography (W-PPG) and electrocardiogram (W-ECG) capabilities. We validated a novel W-PPG algorithm combined with a W-ECG algorithm in a free-living setting, and compared the results to those of a 28-day continuous ECG patch (P-ECG).RESULTS A total of 204 participants completed the free-living study, recording 81,944 hours with both P-ECG and smartwatch measurements. We found sensitivity of 87.8% (95% confidence interval [CI] 83.6%-91.0%) and specificity of 97.4% (95% CI 97.1%-97.7%) for the W-PPG algorithm (every 5-minute classification); sensitivity of 98.9% (95% CI 98.1%-99.4%) and specificity of 99.3% (95% CI 99.1%-99.5%) for the W-ECG algorithm; and sensitivity of 96.9% (95% CI 93.7%-98.5%) and specificity of 99.3% (95% CI 98.4%-99.7%) for W-PPG triggered W-ECG with a single W-ECG required for confirmation of AF. We found a very strong correlation of W-PPG in quantifying AF burden compared to P-ECG (r 5 0.98).CONCLUSION Our findings demonstrate that a novel algorithm using a commercially available smartwatch can continuously detect AF with excellent performance and that confirmation with W-ECG further enhances specificity. In addition, our W-PPG algorithm can estimate AF burden. Further research is needed to determine whether this algorithm is useful in screening for AF in select atrisk patients.

Continuous 24-h Photoplethysmogram Monitoring Enables Detection of Atrial Fibrillation

“…Only 3 patients here received false AF alarms, as low-quality PPG samples were discarded. Our findings also confirmed some earlier results ( Avram et al, 2021 ), that AF can be diagnosed more likely during the night compared to day-time. In resting patients with continuous PPG monitoring, AF was identified with a sensitivity of 81.0%, specificity of 96.4%, and PPV of 86.6% ( Kashiwa et al, 2019 ).…”

“…Photoplethysmography has been previously proven to be able to detect AF ( Bashar et al, 2019 ; Dörr et al, 2019 ; Fan et al, 2019 ; Kashiwa et al, 2019 ; Sološenko et al, 2019 ; Väliaho et al, 2019 ; Avram et al, 2021 ; Väliaho et al, 2021 ), but the signal quality is affected by motion artifacts ( Bashar et al, 2019 ; Sološenko et al, 2019 ; Nabavi et al, 2020 ). Recently, in a continuous AF monitoring study a total of 32.2% of the PPG data was classified as indeterminate, and in line with us more often during the day-time compared to night-time ( Avram et al, 2021 ). In our study, during the daytime, as much as 69.5% of the data recorded from ambulatory patients were discarded.…”

“…PPG-based AF detection in stationary patients from short, a few minute long recordings, can provide AF detection with high sensitivity (93.7-96.4%) and specificity (96.3-99.7%) ( Dörr et al, 2019 ; Fan et al, 2019 ; Väliaho et al, 2019 ; Väliaho et al, 2021 ). With continuous PPG monitoring, sensitivities of 72.0–92.7% and specificities of 97.4–99.7% have been reported ( Bashar et al, 2019 ; Kashiwa et al, 2019 ; Sološenko et al, 2019 ; Avram et al, 2021 ) and discarding poor quality data decreased portion of false positive AF diagnoses ( Bashar et al, 2019 ; Sološenko et al, 2019 ). Despite the problem of a non-interpretable signal and 44.1% of total discarded PPG data, the current study displayed a sensitivity of 95.4%, a specificity of 99.7% and PPV of 97.8% for the detection of AF in a clinical setting in ambulatory patients and 24-h continuous PPG recordings.…”

Machine Learning-Based Optimization of Pre-Symptomatic COVID-19 Detection Through Smartwatch

“…In many cases, a mobile device, e.g. smartwatch, is used as a real-time physiological monitoring system to monitor health and disease status of the patients, such as Li et al 12 , Kruizinga et al 13 , Ding et al 14 , Rodrigues et al 15 , Gruwez et al 16 , Perino et al 17 , Avram et al 18 , and Nasarre et al 19 . Li et al 12 shows the validity of measurements by smartwatches for 6 Minutes Walking Tests and consistency between Walking distances measured by doctors and by smartwatches.…”

“…Ding et al 14 , and Perino et al 17 present the use of a smartwatch and application that demonstrate reasonable accuracy for Atrial fibrillation detection and high usability. Avram et al 18 describes the estimation of the sensitivity and specificity of atrial fibrillation in the user's sinus rhythm in the user's daily life using a smartwatch. Rodrigues et al 15 shows the usability of detecting the position of Alzheimer's patients lost away.…”