Smartphone apps can identify AFib, but false positives give researchers pause

loading...

Smartphone camera apps using photoplethysmography, or PPG, signals to diagnose atrial fibrillation accurately identified patients with the condition, but researchers have concerns about the level of false positives produced by them, according to a review of research published Friday in JAMA Network Open.

Commonly known as irregular heartbeat, AFib, or atrial fibrillation, affects an estimated 6 million people in the United States. However, the condition is believed to be under-diagnosed.

PPG signals assess blood flow through veins and arteries to measure a person's heart rate and, while saying that a high level of false positives gives pause to considering widespread use, researchers think the method offers a significant opportunity for care.

"Smart devices have been proposed to help to detect undiagnosed AFib and aid the management of patients with known AFib," researchers wrote in the study. "We found that all smartphone camera applications individually had a high sensitivity and specificity, and this remained true for the meta-analyzed estimate for all applications collectively."

The group that authored the new analysis, part of the Division of Cardiology at Stanford University School of Medicine, has been investigating the use of technology in diagnosis and management of a variety of heart conditions. UPI reported on their assessment of the Apple Watch in the diagnosis of AFib in November.

For the analysis, the group reviewed the findings of 10 "primary diagnostic accuracy studies" that enrolled a total of 3,852 participants. Collectively, the studies evaluated four separate smartphone camera apps: FibriCheck application, designed by Qompium; Cardiio Rhythm Mobile, designed by Cardiio; Preventicus, from Preventicus; and PULSE-SMART, which is not yet available commercially.

All studies were conducted using iPhones. The oldest studies were published in 2016, while four were published in 2018. The rest were published in 2017 and 2019.

Overall, the authors of the review found the sensitivity and specificity for all applications combined were roughly 94 percent and 96 percent, respectively. Sensitivity is the ability to correctly identify those with a disease, while specificity is the ability of a test to correctly identify those who don't have it.

In addition, the positive predictive value -- the probability that users who get a positive result actually have AFib -- and the negative predictive value -- the probability that those who get a negative result don't have the condition -- for smartphone camera applications in an asymptomatic population age 65 and older were up to 38 percent and 99.8 percent, respectively.

The same figures were up to 39 percent and 99.9 percent, respectively, for people 65 years of age and older with high blood pressure.

"Their modest positive predictive value suggests that these devices will generate a higher number of false positive than true-positive results," the authors noted. "In fact, our model suggests that if these applications detect AF in an asymptomatic person, the result is most likely to be a false-positive. From these results, it would appear premature to use these devices among healthy individuals or to use them to screen an asymptomatic population."