Smartwatch warning: AFib testing reveals ‘high rate’ of false positives for wearable technology using ECG strips

BORDEAUX, France — Health apps on smartwatches and smartphones still lack the ability to accurately detect an irregular heartbeat, according to the largest study on the subject to date. Researchers report that current mobile health technology designed to detect atrial fibrillation produces a “high rate” of false positives and inconclusive results in some patients with certain heart conditions.

While more traditional means of tracking the heart are available, such as B. advanced cardiac monitoring and the use of implantable electronic cardiovascular devices, these devices have limitations, including lack of real-time feedback and short battery life. Meanwhile, new smartphone apps and tools that claim to be able to record an electrocardiogram (ECG) strip and then generate an automated diagnosis are a huge boon to heart patients around the world — but only if they actually work. The study authors set out to determine the accuracy of such mobile technologies, eventually noting that “using these devices in patients with abnormal EKGs is challenging.”

Researchers believe that superior algorithms and improved machine learning could help these tools make more accurate diagnoses in the future.

“Previous studies have validated the accuracy of the Apple Watch for diagnosing atrial fibrillation in a limited number of patients with similar clinical profiles,” explains study leader Marc Strik, MD, PhD, from Bordeaux University Hospital in a press release. “We tested the accuracy of the Apple Watch ECG app to detect atrial fibrillation in patients with a variety of coexisting ECG abnormalities.”

This research involved 734 consecutive hospitalized patients, with each patient undergoing a 12-lead EKG immediately followed by a 30-second Apple Watch recording. The team classified the smartwatch’s automated single-lead ECG-AF (AFib) detections as either “no evidence of atrial fibrillation,” “atrial fibrillation,” or “inconclusive reading.”

1 in 5 patients receive bad information from wearable technology

The smartwatch recordings were sent to an electrophysiologist who performed a blind interpretation and assigned each recording a diagnosis of AF, absence of AF, or diagnosis unclear. Then a second blinded electrophysiologist interpreted 100 randomly selected traces. This helped assess the extent to which the two observers agreed.

In about every fifth patient, the Smartwatch ECF failed to create an automatic diagnosis. Patients with premature atrial and ventricular contractions (PACs/PVCs), sinus node dysfunction, and second- or third-degree atrioventricular block were more likely to receive a false-positive automatic AF detection. In patients with atrial fibrillation, the risk of a false negative curve (missed atrial fibrillation) was higher in patients with ventricular conduction abnormalities (interventricular conduction delay) or pacemaker-controlled rhythms.

In general, the two cardiac electrophysiologists shared a high degree of agreement regarding the distinction between atrial fibrillation and non-atrial fibrillation. The smartphone app successfully identified 78 percent of patients with atrial fibrillation and 81 percent of patients without atrial fibrillation. The electrophysiologists meanwhile discovered 97 percent of the patients with atrial fibrillation and 89 percent without atrial fibrillation.

Patients with PVCs were also three times more likely to receive false positive AFib diagnoses from the smartwatch ECG. In addition, the identification of patients with atrial tachycardia (AT) and atrial flutter (AFL) was quite poor.

“These observations are not surprising since the automated smartwatch detection algorithms are based solely on cycle variability,” notes Dr. Strik and explains that PVCs cause short and long cycles that increase cycle variability. “Ideally, an algorithm would better distinguish between PVCs and AF. Any algorithm that is limited to analyzing cycle variability has poor accuracy in detecting AT/AFL. Machine learning approaches can increase the detection accuracy of smartwatch AF in these patients.”

Smartwatch algorithms “not smart enough yet”

In an accompanying editorial, Andrés F. Miranda-Arboleda, MD, and Adrian Barranchuk, MD, of Kingston Health Science Center, state that this is the first-ever “real-world” study to focus on the Apple Watch as a diagnostic tool for atrial fibrillation concentrated.

“It is of notable importance because it allowed us to learn that the performance of the Apple Watch in diagnosing atrial fibrillation is significantly affected by the presence of underlying ECG abnormalities. In a way, the smartwatch algorithms for detecting atrial fibrillation in patients with cardiovascular disease are not yet smart enough. But they could be soon,” the doctors add.

“With the increasing use of smartwatches in medicine, it is important to know which diseases and ECG abnormalities could influence and alter the smartwatch’s detection of atrial fibrillation in order to optimize the care of our patients,” concludes Dr. strike “Smartwatch detection of atrial fibrillation has great potential but is more challenging in patients with pre-existing heart disease.”

The study appears in Canadian Journal of Cardiology.