Applications
PPG for Atrial Fibrillation (AFib) Detection
PPG-based AF detection identifies atrial fibrillation through the characteristic irregular and irregular rhythm produced by chaotic atrial activity. Algorithms analyze pulse-to-pulse interval irregularity extracted from wrist or finger PPG.
How it works
Schematic summary of this guide. Waveforms are illustrative, not measured data.
Atrial fibrillation produces a hallmark irregularly irregular ventricular response because random conduction of fibrillatory impulses through the AV node creates unpredictable R-R intervals. PPG translates this as highly variable IBI sequences with absent organized P-wave-driven periodicity. Classic AF detection features include RMSSD of IBI, coefficient of variation, entropy measures (sample entropy, approximate entropy), and frequency-domain irregularity metrics applied to rolling 30–60 second windows.
The Apple Heart Study (Perez et al., 2019, NEJM) validated AF detection from Apple Watch irregular pulse notifications: the positive predictive value of an irregular pulse notification for simultaneous AF on a later ECG patch was 0.84 (doi:10.1056/NEJMoa1901183). The study enrolled 419,297 participants and demonstrated real-world feasibility of PPG-based AF screening at population scale.
The Fitbit Heart Study (Lubitz et al. 2022, doi:10.1161/CIRCULATIONAHA.122.060291) reported a positive predictive value of 98.2% for concurrent AF; it used a different definition from the Apple Heart Study, so the two figures are not directly comparable. Apple's irregular rhythm notification feature was granted FDA De Novo classification DEN180042. For how sensitivity, specificity, and study design are treated in that literature, see this evidence review of wearable AF screening study design.
Frequently Asked Questions
How accurate is Apple Watch for AFib detection?
In the Apple Heart Study (Perez et al. 2019), the positive predictive value of an irregular pulse notification for simultaneous AF on an ECG patch was 0.84. A notification still needs ECG confirmation.
Can PPG distinguish AFib from other irregular rhythms?
PPG-based AF detection has high false positive rates for PACs/PVCs, tachycardia, and irregular rhythms from poor signal quality. Rate-corrected entropy metrics and morphological analysis improve specificity but cannot replace ECG confirmation for clinical decisions.
What is the minimum episode length detectable by PPG AF algorithms?
Most validated PPG AF algorithms require 30–300 seconds of continuous AFib to achieve reliable detection. Brief paroxysmal episodes (<30 seconds) are frequently missed, underestimating AF burden compared to implantable loop recorders.