Can a Smartphone Camera Detect Atrial Fibrillation?

Amandeep Singh, M.D.
By Amandeep Singh, M.D. on

An original study published in the Journal of the American Heart Association aimed at finding a novel method for screening patients for atrial fibrillation (AF) using an iPhone. It evaluated the use of the smartphone application ‘Cardiio Rhythm’ in detecting AF by using no physical contact facial photoplethysmographic (PPG) signals.

One of the clinical challenges of AF is the early identification and initiation of stroke prophylaxis. However, the detection of AF can be difficult because it is often asymptomatic and intermittent in duration and thus developing an accessible method for screening is necessary. Thus, the investigators sought to detect a stand-alone application to provide an inexpensive means to screen patients for AF without physical contact. The convenience of a contact-free approach to detect atrial fibrillation could be attractive for community screening and has the potential to be useful for long-distance screening using telemedicine.

A total of 217 inpatients were recruited for 8 months at the Prince of Wales Hospital Cardiology ward in Hong Kong. Two iPhone 6s installed with Cardiio Rhythm application were used for detection of facial and fingertip plethysmography. A 12-Lead ECG was used as a reference tool. The camera could detect subtle beat-to-beat variations of skin color on the basis of changes in reflected light according to the arterial blood volume pulsations. In this study, a positive test screening for AF was defined as either of the following: detection of an irregular heart rhythm in ≥1 PPG measurements or 3 consecutive uninterpretable PPG measurements.

“I believe Cardiio (finger or face) is a user-friendly and potentially cost-effective tool for AF screening but confirmatory ECG is still necessary for diagnosis.”- Dr. Bryan Yan


In the study, the investigators found that the Cardiio Rhythm application identified an irregular pulse in 77 subjects (35.5%) when analyzed on 12 lead ECG. Facial PPG yielded a sensitivity of 94.7% and a specificity of 95.8%. The accuracy was 95.4 %. In comparison, using fingertip PPG, 81 (37.3%) were labeled as an irregular pulse. Fingertip PPG results had a sensitivity of 94.7% and specificity of 93.0 %. The accuracy for fingertip photoplethysmography was 93.5%.  The diagnostic sensitivity of the Cardiio Rhythm fingertip PPG for AF detection was 92.9%, which was higher than that of the AliveCor automated algorithm of 71.4%. Therefore, high sensitivity, low likelihood ratio, and convenience of a contact-free approach were attractive qualities for potential application in large-scale community AF screening. The lower false-positive rate associated with facial PPG(4.2%) suggested that it could serve as a better AF screening method as compared to finger PPG(7.0%).

Emphasizing on the importance of the study, the primary investigator Bryan P. Yan, MD (Prince of Wales Hospital Hong Kong) noted, “This is the first study to demonstrate the feasibility of non-contact AF screening. This has the potential for high-throughput and distant screening.” The authors addressed certain limitations of all PPG-based devices which included the need for a confirmatory ECG for suspected AF. High false-positive rates could limit their use as a mass screening tool because of unnecessary follow-up ECG and consumption of healthcare resources. Detection failures for PPG signals using Cardiio Rhythm application may have occurred because of poor signal quality. Yan et al. stated that plausible causes may have included poor lighting, poor skin perfusion, darker skin tone, low cardiac output state, loss of fingertip contact with the camera, misalignment of the face to the camera, and excessive movement during signal acquisition.  Additionally, in the software, noise filter was not always able to detect poorer signal quality due to variability in pulse volume.

Dr. Yan thinks it is premature to recommend facial Cardiio AF screening at this stage. He said, “I believe Cardiio (finger or face) is a user-friendly and potentially cost-effective tool for AF screening but confirmatory ECG is still necessary for diagnosis.” Acknowledging this limitation, he added, “We are doing further research to refine the facial PPG detection algorithm and method to reduce the recording time and testing in different clinical settings. ”

While the investigators believe that Cardiio Rhythm has immense potential for distant hemodynamic monitoring, they note that patient privacy and consent issues could be a potential concern. “Although we are only analyzing facial PPG signal, patients may be concerned about being filmed and identified without consent. This issue must be resolved before large-scale application of facial PPG detection,” Dr. Yan cautions.

Source: Contact‐Free Screening of Atrial Fibrillation by a Smartphone Using Facial Pulsatile Photoplethysmographic Signals, J Am Heart Assoc. 2018;7:e008585. DOI: 10.1161/JAHA.118.008585

Leave a Reply