AI Atlas Reveals How Body Fat and Muscle Shape Lifelong Health Risks

A new AI-powered study has created a detailed atlas of human body composition using advanced imaging. Researchers analysed over 66,000 MRI scans to map fat and muscle distribution across different ages, sexes and heights. The findings could improve patient care by offering more precise health risk assessments than traditional methods like BMI. The study used deep learning algorithms to measure subcutaneous fat, visceral fat, skeletal muscle and intramuscular fat. These metrics were tracked over time to build reference curves showing how body composition changes with age. Unlike BMI, which does not distinguish between muscle and fat, the AI framework provides a clearer picture of health risks.

Visceral fat, the study found, increases the risk of diabetes by 2.26 times. High levels of intramuscular fat were linked to a 1.54-fold higher chance of major cardiovascular events. Meanwhile, low skeletal muscle mass independently raised the risk of all-cause mortality by 1.44 times.

The AI tool is fully automated and open-source, making it easier for clinics to adopt. It can be applied to routine chest or abdominal CTs and MRIs, allowing for wider use in everyday healthcare. The research offers a more accurate way to assess health risks by focusing on body composition rather than BMI alone. Clinicians can now use the open-source AI framework to analyse scans and identify patients at higher risk of diabetes, heart disease or early death. The findings also provide a baseline for future studies on aging and metabolic health.