Sunday, November 30 | 3:20 p.m.-3:30 p.m. | S5-SSCA02-6 | Room S503
In this presentation, researchers will describe how AI-powered analysis of coronary artery calcium (CAC) on lung CT exams can enable early identification of individuals at risk for diabetes.
As part of the Artificial Intelligence-Cardiovascular Disease (AI-CVD) initiative, presenter Morteza Naghavi, MD, of HeartLung.AI and a multi-institutional research team applied an AI algorithm to analyze baseline CAC scans and up to 19 years of follow-up data for nearly 3,000 normoglycemic, non-obese participants in the Multi-Ethnic Study of Atherosclerosis (MESA).
The software assessed adiposity measures indexed by body surface area, including liver fat, total visceral fat, subcutaneous fat, and pericardial fat. Next, the researchers used Cox proportional hazards regression models to calculate hazard ratios to calculated hazard ratios for incident diabetes. They then determined the importance of the adiposity measures.
Over the 19-year period, 205 (7%) developed diabetes. Data analysis showed that liver fat, total visceral fat, and subcutaneous fat were able to identify normoglycemic, non-obese adults at high risk of future diabetes. Of the various adiposity measures, liver fat was the strongest contributor to the development of type 2 diabetes.
“AI-enabled opportunistic screening in CAC scans can identify individuals at risk for diabetes before the onset of conventional risk factors, enabling early preventive interventions,” the authors wrote.
What else did they discover? Stop by this Sunday afternoon presentation for all of the details.
