Validation of a Novel Contact-free Atrial Fibrillation Screening Method Using an Iphone Camera to Detect Facial Pulsatile Photoplethysmographic Signals

Introduction: We aimed to evaluate a novel method of atrial fibrillation (AF) screening using an iPhone camera to detect and analyze photoplethysmographic (PPG) signals from the face without physical contact by extracting subtle beat-to-beat variations of skin color that reflects the cardiac pulsatile signal similar to heart rate fluctuations. Methods and Results: Facial PPG signal was obtained from 85 hospital in-patients (mean 71.6±13.3 years, 67% male) by having the patients face the front camera of an iPhone 6S before a 12-lead ECG was recorded. Backdrop and background light intensity was standardized during signal acquisition. Three successive recordings of 20 seconds (total 60 seconds) were acquired per patient and analyzed for heart rate regularity by Cardiio Rhythm (Cardiio Inc., Cambridge, MA), a smartphone application. Pulse irregularity in 1 or more of 3 PPG recordings was considered a positive AF screening result. Atrial fibrillation was present in 25 out of the 85 patients (29.4%) based on the 12-lead ECG. The smartphone application demonstrated high sensitivity (0.926, 95% confidence interval (CI) 0.766-0.979) and specificity (0.948%, 95%CI 0.859-0.982) in discriminating an irregular pulse during AF from sinus rhythm compared to 12-lead ECG. The positive and negative likelihood ratios were 17.901 (5.917-54.161) and 0.078 (0.021-0.297), respectively. Conclusions: Our results demonstrated that detection of facial PPG signal to determine pulse irregularity due to AF is feasible. The Cardiio Rhythm smartphone application showed high sensitivity and specificity with low negative likelihood ratio for AF; and the convenience of a contact-free approach is suitable for large-scale community AF screening.