Ultrasound Models Estimate Whole-Body Muscle Mass

New equations incorporating cross-sectional area improve estimation accuracy.

By
Kathryn Wighton
March 26, 2026
4 min

MDSpire News

Objective:

To develop and validate ultrasound-derived equations for estimating whole-body muscle mass, emphasizing the significance of MRI as the reference standard.

Key Findings:

The combined model (Model 5) achieved an adjusted R² of 0.948 and a standard error of estimate (SEE) of 1.6 kg in the development group, indicating high accuracy.
In cross-validation, Model 5 showed a bias of −0.03 kg and an intraclass correlation coefficient of 0.965, demonstrating reliability.
The most accurate equation included 10 variables, while a practical equation with 6 variables had an adjusted R² of 0.927 and a SEE of 2.0 kg, reflecting a trade-off between accuracy and measurement burden.

Interpretation:

Models combining muscle thickness and CSA measurements improved prediction accuracy compared to those using muscle thickness alone, highlighting their practical implications.

Limitations:

The equations are applicable only to healthy Caucasian populations, necessitating further research in diverse groups.
Measurements of muscles with complex morphology may be prone to inter-rater variation.
MRI limitations required unilateral measurements in some participants.

Conclusion:

The addition of CSA measurements to muscle thickness models enhances accuracy in estimating whole-body muscle mass, underscoring their importance in practical applications.