AI Analysis of Pericardial Fat Refines Long-Term Heart Disease Risk

Missed or underestimated risk delays preventive therapy and increases the chance of heart attack and stroke. To help address this challenge, researchers have now developed an artificial intelligence approach that extracts additional prognostic data from a routine computed tomography scan. The method quantifies fat around the heart, known as pericardial adipose tissue, to sharpen risk estimates without new imaging.

Mayo Clinic investigators evaluated an AI–derived measurement of pericardial adipose tissue generated from standard coronary artery calcium scans. The approach analyzes existing images to quantify fat surrounding the heart. The resulting volume measurement is assessed as an independent risk marker and as an adjunct to established calculators.

In a longitudinal study of nearly 12,000 adults followed for approximately 16 years, researchers compared pericardial fat volume with two common risk tools. These comparators were the American Heart Association PREVENT equation, which incorporates traditional factors, and the coronary artery calcium score. Analyses examined predictive value when each method was used alone and in combination.

Pericardial fat volume independently predicted cardiovascular events and improved the overall accuracy of long-term risk prediction when added to the PREVENT equation and the coronary artery calcium score. The enhancement was most pronounced among patients categorized as low risk by standard models. Nearly 10% of participants developed cardiovascular disease during follow-up, and higher fat volume conferred elevated risk across all calcium score strata.

Because the measurement is derived from scans many patients already receive, the strategy could augment prevention workflows without additional testing. Findings were presented at the 2026 American College of Cardiology Scientific Session with simultaneous publication in the American Journal of Preventive Cardiology. The team noted that further studies are needed to define how pericardial fat measurement should be incorporated into routine clinical decision-making.

"Pericardial fat has been recognized as a marker of cardiovascular risk, but this study shows how we can now measure it automatically and use it to meaningfully improve risk prediction, especially in patients at borderline or intermediate risk where clinical decisions are often less clear. This opens the door to more personalized prevention strategies," said Zahra Esmaeili, first author and researcher in the Department of Cardiovascular Medicine at Mayo Clinic.

"Because this measurement comes from imaging that many patients are already receiving, it represents a practical and scalable way to enhance cardiovascular risk assessment. It could help clinicians intervene earlier and more effectively," said Francisco Lopez-Jimenez, M.D., a preventive cardiologist and co-director of the AI in Cardiology program at Mayo Clinic.