The Mayo Clinic in the US has conducted a study to evaluate the use of artificial intelligence (AI) technology in detecting the signs of atrial fibrillation (AF) or irregular heart rhythms in an EKG.
According to the findings, the technology was able to detect AF even when the heart rhythm was normal at the time of the test.
The researchers believe that the AI-enabled EKG will be able to identify recent AF that occurred without symptoms or an imminent AF, thus enabling better treatment.
EKG is widely used to detect AF, which may not occur during the conventional ten-second, 12-lead test. Long-term monitoring approaches such as a loop recorder are costly and require a procedure.
When left untreated, the condition can lead to stroke, heart failure and other cardiovascular diseases.
To train the new detection technology, the researchers used 450,000 EKGs out of the more than seven million EKGs available in the Mayo Clinic digital data vault.
The AI was trained to detect subtle differences in a normal EKG that would point to changes in heart structure caused by the condition.
Using normal-rhythm EKGs from 36,280 patients, including 3,051 with known atrial fibrillation, the team tested the AI. It was observed that the AI-enabled EKG demonstrated a 90% accuracy in correctly detecting subtle AF patterns.
Mayo Clinic Cardiovascular Medicine department chair Paul Friedman 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.
“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.”
Friedman added that the AI technology can be used with a smartphone or watch, allowing its availability on a large scale.