Researchers have developed a new artificial intelligence algorithm that measures coronary plaque buildup and may help predict heart attacks.
Plaque buildup causes the arteries to narrow and makes it difficult for the blood to flow, increasing the likelihood of a heart attack. Although coronary computed tomography angiography (CTA) can estimate how much a patient’s arteries have narrowed, there has not been a simple, automated, and rapid way to measure the plaque visible in a CTA image, until now.
In the new study, researchers analyzed CTA images from nearly 2,000 people who had undergone a coronary CTA at 11 sites worldwide, including the United States. The researchers trained their AI algorithm to measure plaque based on previous analyses of coronary CTA images that had been studied by trained doctors.
The researchers found that the measurements made by their AI algorithm accurately predicted heart attack risk within five years for most of the people who were part of their multicenter trial. The AI algorithm can quantify plaque from CTA images in just five to six seconds, a process that normally takes an expert 25 to 30 minutes to complete, the researchers said.
The study, funded in part by the NHLBI, appeared in The Lancet Digital Health.