A novel imaging-based "sarcopenic obesity index" — combining muscle mass and body fat measurements from cardiac MRIs — predicted heart failure, cardiovascular death, and all-cause mortality better than BMI or waist-based measures. Developed using deep learning on over 55,000 UK Biobank scans, the index also revealed genetic links between muscle-fat imbalance and heart disease. It could reshape how clinicians assess cardiac risk.
Researchers have developed a new imaging-based tool called the sarcopenic obesity index that may change how doctors assess cardiovascular risk. Built using a deep learning algorithm applied to over 55,000 cardiac MRI scans from the UK Biobank, the index combines body weight with 3D measurements of pectoralis major muscle mass to capture the dual burden of excess fat and low muscle — a combination traditional metrics like BMI simply can't detect.
The index was linked to adverse changes in heart structure, including signs of diffuse myocardial fibrosis, and outperformed BMI, waist-to-hip ratio, and waist-to-height ratio in predicting cardiac outcomes. Genetic analyses also uncovered 16 sarcopenic obesity-related gene loci, several of which overlap with known heart failure and cardiomyopathy genes.
By the Numbers:
Why it matters: BMI has long been criticized for missing the nuances of body composition. This index offers a more precise, imaging-derived way to identify high-risk patients — potentially enabling earlier, targeted interventions for those with sarcopenic obesity.
