AI leverages ECG data to predict patient’s overall health status

Register now

Using electrocardiogram data, Mayo Clinic researchers have developed a convolutional neural network that is able to determine physiologic age, a measure of patients’ overall body function and health status.

The artificial intelligence accurately determined a patient’s gender with 90 percent accuracy and chronological age group with 72 percent accuracy. However, what was a surprising discovery for researchers was how discrepancies between AI-predicted age and chronological age ended up serving as a physiological measure of health.

While the AI was trained to predict a person’s age and self-reported sex using standard 12-lead ECGs, it also estimated a patient's chronological age as higher after experiencing adverse health situations—such as heart attack, low ejection fraction and coronary artery disease—and lower age if they experienced few or no adverse events.

“When the convolutional neural network-predicted age exceeds a patient’s actual age by at least 7 years, there is a higher incidence of cardiovascular comorbidities, potentially suggesting that the convolutional neural network-predicted age from 12-lead ECGs may correlate with physiological health,” according to a study, published on Tuesday in the journal Circulation: Arrhythmia and Electrophysiology.

Also See: AI predicts long-term mortality by analyzing chest X-rays

The study’s authors contend that the ability of an AI algorithm to determine physiological age—with further validation in well-characterized, prospectively followed cohorts—has the potential to serve as a rapid measure of overall health status.

“While physicians already consider whether a patient ‘appears [their] stated age’ as part of their baseline physical examination, the ability to more objectively and consistently assess this may impact healthcare on multiple levels,” said study author Suraj Kapa, MD, assistant professor of medicine and director for Augmented and Virtual Reality Innovation at Mayo Clinic.

“Being able to more accurately assess overall health status may help doctors determine which patients they should examine further to determine if there are asymptomatic or currently silent diseases that could benefit from early diagnosis and intervention,” added Kapa. “For people at large, an AI-enhanced electrocardiogram could better show there may be something going on such as a new health issue or comorbid condition that they were otherwise unaware of.”

The Mayo Clinic study was funded using institutional funds for data collection and statistical analyses.

For reprint and licensing requests for this article, click here.