Novel Body Shape Descriptors for Abdominal Adiposity Prediction Using Magnetic Resonance Images and Stereovision Body...

Area of Science:

• Biomedical imaging
• Anthropometry
• Body composition analysis

Background:

• Abdominal adiposity, particularly visceral adipose tissue (VAT), is a significant health concern linked to metabolic diseases.
• Accurate measurement of abdominal fat is crucial for risk assessment and management.
• Traditional methods may be invasive or lack detailed body shape information.

Purpose of the Study:

• To develop novel shape descriptors from 3D body images.
• To establish predictive models for abdominal adiposity using these descriptors.
• To validate the efficacy of 3D shape descriptors in estimating visceral adipose tissue (VAT) and subcutaneous adipose tissue (SAT).

Main Methods:

• Recruited 66 men and 55 women for 3D whole-body and abdominal MRI scans.
• Quantified VAT and SAT volumes using an automated algorithm on MRI data.
• Developed and validated multiple regression models using 3D body shape descriptors, age, and BMI to predict VAT and SAT.

Main Results:

• Thirteen body shape descriptors showed significant correlations (P < 0.01) with abdominal adiposity.
• Optimal predictive models for VAT and SAT were established, with separate equations for men and women.
• The developed models demonstrated effective prediction of abdominal fat volumes.

Conclusions:

• Novel body shape descriptors derived from 3D imaging provide an effective means to predict abdominal adiposity.
• This approach offers a non-invasive and potentially more accessible method for assessing body fat distribution.
• 3D body imaging combined with shape analysis holds promise for clinical applications in obesity assessment.