Behind The Science: Radiomic Profiling of Epicardial Fat for Heart Failure Prediction | JACC

In this episode of Behind the Science, host Adith Arun sits down with Dr. Evangelos Oikonomou to discuss groundbreaking research published in JACC: "Early Prediction of Heart Failure from Routine Cardiac CT Using Radiomic Phenotyping of Epicardial Fat." Dr. Oikonomou walks us through how AI-powered radiomic analysis of epicardial adipose tissue, the fat layer directly surrounding the heart, can serve as an early biomarker for heart failure risk, even in patients with no prior diagnosis. The study draws on over 60,000 coronary CT angiography cases from nine UK sites (the ORFNI consortium) to build and validate a fully automated risk scoring model. Topics covered: Why BMI falls short as a cardiovascular risk measure What epicardial fat reveals about myocardial disease How radiomics extracts hidden phenotypic features from CT images Model training, validation, and multi-site generalizability How the FRP (Fat Radiomic Profile) score is orthogonal to existing tools like PREVENT Implications for GLP-1 therapies and obesity-directed treatment eligibility The future of longitudinal monitoring and prospective enrichment studies This research represents a major step toward extracting untapped prognostic value from CT scans already being performed in clinical practice.