Evaluating waist-to-hip ratio in youth using frequency-modulated continuous wave radar and machine learning

Area of Science:

• Biomedical Engineering
• Medical Imaging
• Pediatric Health

Background:

• Waist-to-hip ratio (WHR) is crucial for predicting cardiometabolic diseases.
• Traditional tape measurements for WHR in children are uncomfortable and variable.
• Accurate central obesity assessment is vital for pediatric health monitoring.

Purpose of the Study:

• To develop a non-invasive frequency modulated continuous wave (FMCW) radar system for precise WHR measurement.
• To evaluate the accuracy of radar-based WHR assessment against traditional methods.
• To assess the utility of radar technology for central obesity risk stratification in adolescents.

Main Methods:

• Developed an FMCW radar system for non-invasive body measurement.
• Utilized point cloud generation and convolutional neural networks for radar data analysis.
• Compared radar-derived WHR measurements with clinician-obtained tape measurements in 100 participants (aged 7-18).

Main Results:

• Demonstrated strong agreement between radar-based and traditional WHR measurements (ICC=0.83).
• Achieved 82% accuracy in classifying participants into abdominal obesity risk groups.
• The radar system proved reliable for routine central obesity monitoring.

Conclusions:

• FMCW radar offers a precise, non-invasive, and convenient method for WHR measurement in children and adolescents.
• This technology enhances privacy and comfort, facilitating widespread clinical and non-clinical application.
• Radar-based WHR assessment is a viable tool for monitoring central obesity and related health risks.