AI-CVD-HF

A heart failure risk prediction model based on coronary artery
calcium scans compared with PREVENT-HF

The AI-CVD-HF model introduces a novel imaging-based approach for predicting heart failure (HF) by integrating AI-derived imaging biomarkers extracted from Coronary Artery Calcium (CAC) scans with basic demographic variables, including age and sex. The model outperformed the baseline version of the recently introduced PREVENT-HF risk score, demonstrating superior discrimination and calibration while maintaining robust and consistent predictive performance across different age groups, sexes, diverse racial and ethnic populations, and an independent external validation cohort. These results highlight the model’s strong potential for generalizability and support its clinical applicability for opportunistic HF risk assessment using routine chest CT scans. However, prospective randomized clinical trials and comprehensive cost-effectiveness analyses are still needed before the model can be recommended for widespread clinical implementation.