A new artificial intelligence tool has been developed that researchers say may help predict heart attacks in patients before they happen. 

Investigators from Cedars Sinai Medical Center say they have developed a novel deep-learning tool and artificial intelligence algorithm that may be able to predict the risk of a heart attack or other cardiac event based on patient health data and by interpreting heart images.

“[W]e could both predict the chance of cardiac events — like death, heart attack, or the need for urgent treatment of the heart vessels — and show how the likelihood of these adverse events changes over time,” said Piotr Slomka, PhD, director of Innovation in Imaging at Cedars-Sinai and a research scientist in the Division of Artificial Intelligence in Medicine and the Smidt Heart Institute at Cedars Sinai, in a news release.

For their study, the researchers examined a cohort of 20,400 patients following a major adverse cardiovascular event (MACE) for a median of 4.4 years. It also observed patients from that group who eventually died from cardiac events or other causes over that period. 

Using these data, researchers developed an AI tool which is able to predict a patient’s likelihood of suffering a heart attack or cardiac event based on their risk factors, clinical history and images of the heart. 

Predictions are then displayed in an easy-to-understand graph format for medical professionals and patients to view, which indicates a patient’s risk for a heart attack, cardiac surgery or death over the period of several years. This allows medical professionals to suggest changes in a patient’s lifestyle or health to reduce the likelihood of a cardiac event. 

“Doctors and patients can use these graphs to track how risk changes over time and to identify individual risk factors,” Slomka said in the release. “They can also interactively modify certain risk factors to see how it impacts a patient’s particular risk.”

Slomka and his team plan to test these tools in clinical trials at Cedars Sinai in the near future to test their broader applicability in clinical settings. “Our ultimate goal is to offer such interactive tools online if images and clinical data are uploaded, he said.  

The study was published in the May 1 issue of the npj Digital Medicine journal.