Skip to main content

Analysis

AI IDs Afib

By John Commins  
   August 02, 2019

AI was taught to identify subtle differences in a normal EKG that would indicate changes in heart structure caused by afib.

Artificial intelligence has proved to be adept at detecting recent or impending atrial fibrillation, even if symptoms aren't noticeably apparent, new research shows.

In a study published in The Lancet, Mayo Clinic researchers found that AI can detect the signs of an irregular heart rhythm in an EKG, even if the heart is in normal rhythm at the time of a test.

Study senior author Paul Friedman, MD, chair of the Department of Cardiovascular Medicine at Mayo Clinic, says the findings could prove critically important when treating stroke victims.  

"When people come in with a stroke, we really want to know if they had AF in the days before the stroke, because it guides the treatment," Friedman said in comments accompanying the study.

"Blood thinners are very effective for preventing another stroke in people with AF. But for those without AF, using blood thinners increases the risk of bleeding without substantial benefit," he said. "That's important knowledge. We want to know if a patient has AF."

Using 450,000 EKGs of the over 7 million EKGs in the Mayo Clinic digital data vault, the researchers taught AI to identify subtle differences in a normal EKG that would indicate changes in heart structure caused by afib. The changes were not detectable without AI.

The researchers then tested the AI on normal-rhythm EKGs of 36,280 patients, of whom 3,051 were known to have afib and found that the AI-enabled EKG correctly identified afib with 90% accuracy, the study said.

In addition to improving diagnoses for afib, Friedman said AI-guided EKGs can be processed using a smartphone or watch, making it readily available on a large scale.

"An EKG will always show the heart's electrical activity at the time of the test, but this is like looking at the ocean now and being able to tell that there were big waves yesterday," Friedman said.

"AI can provide powerful information about the invisible electrical signals that our bodies give off with each heartbeat — signals that have been hidden in plain sight," he said.  

“An EKG will always show the heart's electrical activity at the time of the test, but this is like looking at the ocean now and being able to tell that there were big waves yesterday.”

John Commins is a content specialist and online news editor for HealthLeaders, a Simplify Compliance brand.


KEY TAKEAWAYS

Using 450,000 EKGs, the researchers taught AI to identify subtle differences in a normal EKG that would indicate changes in heart structure caused by afib.

The researchers then tested the AI on normal-rhythm EKGs of 36,280 patients, of whom 3,051 were known to have afib and found that the AI-enabled EKG correctly identified afib with 90% accuracy.


Get the latest on healthcare leadership in your inbox.