Skip to content

Abstract 16949: AI-Derived Biological Cardiovascular Age from Coronary CT Angiography Predicts 1-Year Major Adverse Cardiac Events

Nov 2026 · Circulation · 0 citations

Abstract

Background Chronological age incompletely reflects cardiovascular biological aging and associated clinical risk. We evaluated whether an artificial intelligence (AI)-derived biological age from coronary CT angiography (CCTA) predicts short-term major adverse cardiac events (MACE). Methods An ordinal machine-learning model integrating automated CCTA-derived plaque, stenosis, and cardiac structural features was developed in a multisite cohort (n=7,843) and tested in an independent cohort with 1-year follow-up (n=1,672). SHAP-guided feature reduction identified a parsimonious 5-feature model incorporating left atrial volume, left ventricular epicardial volume, right ventricular volume, calcified plaque volume, and stenosis severity. A bias-corrected biological age gap (predicted minus chronological age) defined accelerated cardiovascular aging. Results Accelerated cardiovascular aging (age gap ≥2 years) identified patients with significantly higher 1-year MACE compared with non-accelerated aging (20.5% vs 5.5%, p<0.001) (Figure), corresponding to a 3.7-fold increase in risk. In comparison, Agatston-derived arterial age gap demonstrated weaker discrimination (17.6% vs 9.4%, p<0.05; 1.9-fold increase). Concordance analysis demonstrated complementary risk stratification between AI-derived biological age and arterial age gap. Patients with both markers elevated had the highest event rate (23.5%), whereas patients with neither elevated had the lowest risk (3.0%). Intermediate event rates were observed when only one marker was elevated. Conclusions AI-derived biological cardiovascular age from CCTA provides a scalable multisystem biomarker integrating cardiac remodeling and atherosclerotic burden. This approach improves short-term risk stratification beyond calcium-based vascular aging metrics and may enable opportunistic identification of high-risk patients undergoing routine coronary CTA.

View source

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.