AI Can Predict Future Heart Attacks By Analyzing CT Scans

An artificial intelligence platform developed by an Israeli startup can reveal whether a patient is at risk of a heart attack by analyzing their routine chest CT scans. 

Results from a new study testing Nanox.AI’s HealthCCSng algorithm on such scans found that 58 percent of patients unknowingly had moderate to severe levels of coronary artery calcium (CAC) or plaque. 

CAC is the strongest predictor of future cardiac events, and measuring it typically subjects patients to an additional costly scan that is not normally covered by insurance companies. 

HealthCCSng was tested on the scans of 326 patients at Beilinson Hospital in Petah Tikva, central Israel.

Among the patients who participated in the study, 101 (31 percent) had severe CAC, 88 (27 percent) had moderate CAC, and 137 (42 percent) had low CAC. 

“The patients in this study received routine CT scans that had nothing to do with a cardiac concern,” said Professor Ran Kornowski, Director of the Cardiology Center at Beilinson Hospital, who led the study.

“Nanox’s AI technology can enable physicians to route these unsuspecting individuals with high CAC levels to the appropriate care pathways and treatment,” he said.

“While the study’s findings were staggering, we are encouraged by the important role AI can play in early risk identification and prevention of cardiac events.”