AI Atlas Reveals How Body Fat and Muscle Predict Health Risks Better Than BMI

Researchers have developed an AI-powered atlas of human body composition using advanced imaging. The project analysed whole-body MRI scans from over 66,000 individuals. Its findings challenge the reliance on BMI as a key health indicator by revealing more precise risk factors linked to fat and muscle distribution. The study used open-source AI to extract detailed body composition metrics from MRI scans with minimal human input. This automated framework can also be applied to routine chest or abdominal CTs and MRIs, making it scalable for clinical use.

Key discoveries include a 1.54-fold higher risk of major cardiovascular events in individuals with elevated intramuscular fat. Low skeletal muscle mass was independently linked to a 1.44-fold increase in all-cause mortality. Additionally, visceral fat showed a strong connection to diabetes risk, with a 2.26-fold rise in likelihood. The research also produced reference curves showing how body composition changes with age. These trajectories provide clearer benchmarks than BMI for assessing health risks across different life stages.

The atlas offers a more accurate way to predict health risks by focusing on fat and muscle metrics rather than BMI alone. Clinicians can now use automated AI tools to assess body composition from standard imaging. This approach may improve early detection of conditions like diabetes and cardiovascular disease in routine medical practice.