AI gains ground in cardiac health While health tech watchers question how much artificial intelligence really moves the needle in medicine today, a pair of new studies point to its promise in a specific area: predicting risk of heart issues. One study, conducted by scientists at Cedars Sinai, found that a deep learning tool could help identify patients at risk of a heart attack by scanning 3D heart images for plaque buildup. Until recently, researchers said, there hasn’t been a quick or automated way to measure the plaque that shows up in those images. The study found the tool’s measurements matched up to expert readers’ analysis of the images, as well as to more invasive tests. While there’s more to study, “it’s possible we may be able to predict if and how soon a person is likely to have a heart attack based on the amount and composition of the plaque imaged with this standard test,” co-author Damini Dey said in a release about their study, which was published in The Lancet Digital Health. In another study — this one just out in the Journal of the American College of Cardiology — researchers used features within an electronic health record to build a new tool predicting patients’ risk of coronary artery disease. That tool increased the prediction and reclassification of patients for coronary artery disease, suggesting that it could be useful for assessing short-term risk at large health systems, they wrote. | 
(American Medical Association)
No comments