ESPE Abstracts

Background: Central obesity in children is a significant risk factor for cardiovascular diseases, potentially more critical than BMI-defined obesity, and can persist into adolescence and adulthood. Measuring the Waist-to-Hip Ratio (WHR), an indicator of central obesity and associated health risks, typically requires direct measurement with a tape, which can be inconvenient and socially unacceptable. Therefore, non-contact anthropometric techniques to measure both body mass index (BMI) and WHR are necessary for routine assessment in clinical settings.

Objective: This study proposes a non-contact approach for measuring WHR, BMI, and analyzing abdominal obesity using frequency modulated continuous wave (FMCW) radar.

Method: We enrolled 131 participants (54 girls and 77 boys) aged 7-18 from the pediatric endocrinology outpatient clinic at Hanyang University Hospital in Seoul, South Korea. Patients were classified into normal, overweight, obese, and severely obese groups based on BMI, and 15 patients from each group were randomly selected, totaling 60 patients. Based on WHR values, participants were categorized into three abdominal obesity risk groups: high risk (WHR ≥ 0.9 for males and ≥ 0.85 for females), moderate risk (WHR < 0.85 for males and < 0.80 for females), and low risk (WHR > 0.85 and ≤ 0.9 for males and > 0.80 and ≤ 0.85 for females). Data were collected using FMCW radar while participants stood upright for 5 seconds, with waist and hip circumferences measured by a medical specialist. Images were generated from the data collected by the radar, and WHR was measured using a convolutional neural network (CNN). Agreement between actual measurements and radar measurements was evaluated using correlation analysis and Bland-Altman plots.

Results: For the WHR measured using radar data, the intra-class correlation coefficient (ICC) was 0.74, with a p-value less than 0.05, confirming a significant correlation between actual body measurements and those obtained through radar technology. The accuracy, recall, and precision of the categorized abdominal obesity risk based on the measured WHR and the actual measurements were 0.84, 0.83, and 0.83, respectively, demonstrating a high level of correlation.

Conclusion: The experimental results confirmed a high correlation between the WHR and BMI measured by radar and the actual measurements, and the effective identification of abdominal obesity risk through radar-based WHR analysis. In future research, we will recruit more participants and further examine the anthropometric characteristics of participants in each group.