Related Experiment Video
Updated: Apr 6, 2026
06:48
Author Spotlight: Advancements in 3D Optical Imaging for Comprehensive Body Composition Assessment in Modern Research
Published on: June 7, 2024
2.3K
Accurate body composition measures from whole-body silhouettes
Bowen Xie1, Jesus I Avila1, Bennett K Ng1
1Department of Radiology and Biomedical Imaging, University of California, San Francisco, California 94115-0628.
Medical Physics
|August 3, 2015
Summary
Body composition parameters like fat mass index can be estimated from children's whole-body silhouettes using active shape modeling. Combining silhouette data with demographics improves accuracy for assessing nutritional status and muscle development.
Area of Science:
- Biometrics and Anthropometry
- Medical Imaging and Analysis
- Pediatric Health
Background:
- Obesity and related conditions like diabetes are significant global health concerns affecting millions.
- Accurate assessment of body composition, including fat mass index (FMI), fat-free mass index (FFMI), and percent body fat, is crucial for evaluating nutritional status and muscle development.
- Traditional methods for body composition analysis can be resource-intensive and may not be suitable for widespread screening.
Purpose of the Study:
- To investigate the feasibility of using frontal whole-body silhouettes and active shape modeling (ASM) to accurately measure body composition parameters in children.
- To determine if ASM techniques applied to silhouette data can provide reliable estimates of FMI, FFMI, and percent fat mass.
- To compare the predictive accuracy of ASM-based models with traditional demographic-based models.
Main Methods:
- Generated binary silhouette images from dual-energy x-ray absorptiometry (DXA) scans of 200 healthy children (ages 6-16).
- Utilized ASM to describe shape variations and identify principal components (modes) of shape.
- Developed predictive models for FMI, FFMI, and percent fat using stepwise linear regression, comparing shape-based models with demographic-based models (age, sex, height, weight, BMIZ).
Main Results:
- Active shape modeling explained 95% of shape variation using 26 modes.
- Shape-based models and demographic-based models showed similar prediction accuracy for most body composition variables.
- Combining silhouette shape data with demographics significantly improved the prediction accuracy for FMI, FFMI, and percent fat in both boys and girls, achieving adjusted R-squared values up to 0.95.
Conclusions:
- Whole-body silhouettes, analyzed with ASM, offer a viable method for estimating body composition parameters (FMI, FFMI, percent fat) in children.
- These findings suggest the potential for developing non-invasive, camera-based methods for body composition assessment, similar to those used in mobile devices.
- This approach could facilitate more accessible and frequent monitoring of pediatric nutritional status and body composition.
