Researchers aimed to find a screening method using computed tomography calcium scoring (CTCS) to assess the risk of heart failure (HF) in patients, particularly focusing on those with type 2 diabetes.
They analyzed CTCS scans from nearly 2,000 patients and applied deep learning to create models that predict HF risk based on radiomic features of epicardial adipose tissue (EAT) and calcifications.
The study found that CTCS-based models, especially those using fat-omics for non-diabetic patients and calcium-omics for diabetic patients, significantly outperformed traditional clinical prediction methods in forecasting incident HF.
Recent studies have focused on measuring epicardial adipose tissue (EAT) to predict major cardiovascular events, but traditional metrics have shown limited effectiveness.
This study aimed to develop advanced EAT features called "fat-omics," which capture more detailed aspects of EAT's role in cardiovascular health, enhancing MACE prediction.
Through testing a cohort of 400 patients, the novel 15-feature fat-omics model significantly outperformed traditional measures, showing better risk stratification for cardiovascular events.
Background: Recent studies have used basic epicardial adipose tissue (EAT) assessments (e.g., volume and mean HU) to predict risk of atherosclerosis-related, major adverse cardiovascular events (MACE).