Deep learning-based biological age estimation from magnetic resonance imaging predicts cardiometabolic outcomes in the general population

Motivation: Little research has been done into how the body's anatomy can indicate aging. Goal(s): The goal of this study is to develop and validate new measures of biological age based on an image of an individual's internal anatomy. Approach: We use deep learning to estimate biological age from fat and muscle in abdominal MRI (we refer to as MRI-Age) Results: Deep learning-derived MRI-Age predicts cardiometabolic outcomes in the general population beyond chronological age, body composition measures, and cardiometabolic risk factors. Impact: Individuals at high MRI-Age could benefit from personalized prevention strategies, lifestyle interventions, and treatment planning.