Abstract

This work combines Dixon-based quantitative MRI and automated segmentation to measure fat fraction (FF) and lean normalized volume (LNV) across seven core muscles, and to derive composite core scores (FFcore and LNVcore). Highly active cyclists (n = 84) were compared with physically inactive volunteers (n = 85), with analyses adjusted for age, sex, and BMI.

Cyclists showed lower FF and higher LNV across muscles and composite scores. Age was a significant predictor of FF for all muscles and FFcore, while activity group was a significant predictor for most muscles. For LNV, activity group, sex, and BMI were significant predictors across all muscles and LNVcore. The results highlight muscle-specific activity-related differences and identify FFcore as a broadly sensitive summary metric across activity and demographic/body-composition factors.