Are smartwatch health apps smart enough to detect atrial fibrillation? — Science Daily

“Previous studies have confirmed the accuracy of the Apple Watch in the diagnosis of AF in specific patients with similar clinical profiles,” said lead investigator Marc Strick, MD, PhD, LIRYC Institute, Bordeaux University Hospital, Bordeaux, France. “We tested the accuracy of the Apple Watch ECG app for detecting AFF in patients with various coexisting ECG abnormalities.”

The study included 734 consecutive hospitalized patients. Each patient has a 12-lead ECG, immediately recording a 30-second Apple Watch. The smartwatch’s automated single-lead ECG AF detectors are classified as “no signs of atrial fibrillation,” “atrial fibrillation” or “null readings.” Smartwatch recordings were provided to an electrophysiologist who performed blinded interpretation, assigning each finding a diagnosis of “AF”, “absence of AF”, or “diagnosis unclear”. A second blinded electrophysiologist interpreted 100 randomly selected traces and determined the extent to which the observers agreed.

In approximately one in five patients, the smartwatch failed to perform an automatic ECG diagnosis. Patients with premature atrial and ventricular contractions (PACs/PVCs), sinus node dysfunction, and second- or third-degree atrioventricular-block are more likely to have false-positive autologous AF. For patients in AF, patients with ventricular conduction delay (interventricular conduction delay) or patients managed with an implantable pacemaker are at increased risk of false negative findings (missed AF).

Cardiac electrophysiologists had a high agreement on the difference between AF and AF. The smartphone application correctly identified 78% of patients in AF and 81% of patients not in AF. Electrophysiologists identified 97% of patients who were in AF and 89% who were not.

PVC patients were three times more likely to have false-positive AF diagnoses from smartwatch ECG, and identification of patients with atrial tachycardia (AT) and atrial flutter (AFL) was poorer.

“These observations are not surprising given that smartwatch automated detection algorithms are based solely on cycle variability,” explains Dr. Strick, adding that PVCs create both short and long cycles, which increase cycle variability. “Ideally, an algorithm would be better able to discriminate between PVC and AF. Any algorithm limited to cycle dynamics analysis would have poor accuracy in identifying AT/AFL. Machine learning approaches could increase the accuracy of a smartwatch in identifying AF in these patients.”

In an accompanying editorial, Andres F. Miranda-Arboleda, MD, and Adrian Baranchuk, MD, Department of Cardiology, Kingston Health Sciences Center, Kingston, ON, Canada, present this first “real-world” study using the Apple Watch as an AF diagnostic tool.

“This is of great importance, because it allowed us to study the performance of the Apple Watch in the diagnosis of AF, because the presence of underlying ECG abnormalities is significantly affected. In a way, the algorithms of the smartwatch detect AF in patients. Cardiovascular conditions are not smart enough. But they may be soon,” said Dr. Miranda-Arboleda and Dr. Baranchuk.

“With the increasing use of smartwatches in medicine, it is important to know which health conditions and ECG abnormalities influence and modify the detection of AF by smartwatches in order to optimize the care of our patients,” said Dr. Strick. “Smartwatch detection of AF has great potential, but is more challenging in patients with pre-existing heart disease.”

Source of history:

Materials Presented by other. Note: Content may be edited for style and length.