Related Experiment Video
Updated: Jan 18, 2026
13:35
Segmentation and Linear Measurement for Body Composition Analysis using Slice-O-Matic and Horos
Published on: March 21, 2021
11.6K
Construct prediction models for low muscle mass with metabolic syndrome using machine learning
Yanxuan Wu1,2, Fu Li1,2, Hao Chen1,2
1Neck-Shoulder and Lumbocrural Pain Hospital of Shandong First Medical University, Shandong First Medical University & Shandong Academy of Medical Sciences, Jinan, China.
Plos One
|September 9, 2025
Summary
This study developed a machine learning model to identify low muscle mass in young adults with metabolic syndrome. Key predictors include height, gender, waist circumference, thigh length, and alkaline phosphatase, aiding early sarcopenia screening.
Area of Science:
- Gerontology
- Metabolic Health
- Biostatistics
Background:
- Metabolic syndrome (MetS) and sarcopenia are significant global health issues, particularly concerning in younger populations.
- Co-occurrence of MetS and sarcopenia elevates mortality risk, necessitating effective identification strategies.
- Current methods for detecting low muscle mass in MetS patients are underdeveloped.
Purpose of the Study:
- To develop and validate a machine learning model for identifying low muscle mass in young MetS patients.
- To identify key predictors associated with low muscle mass in this demographic.
- To create an accessible tool for sarcopenia screening in at-risk individuals.
Main Methods:
- Utilized data from 2,467 MetS patients (aged 18-59) from the 2011-2018 NHANES.
- Employed Least Absolute Shrinkage and Selection Operator (LASSO) regression for feature selection.
- Constructed and evaluated nine machine learning models, including Logistic Regression, using metrics like AUC and F1-score, and interpreted predictors with SHAP values.
Main Results:
- The Logistic Regression model demonstrated superior performance with an AUC of 0.925, an F1-score of 0.87, and specificity of 0.89.
- Identified height, gender, waist circumference, thigh length, and alkaline phosphatase (ALP) as significant predictors of low muscle mass.
- SHAP values were used to interpret the contribution of each predictor.
Conclusions:
- An interpretable machine learning model was successfully developed to pinpoint risk factors for low muscle mass in young MetS patients.
- The findings provide a foundation for improved sarcopenia screening and management in this population.
- A web-based tool was created to enhance the practical application of these findings for early detection.
