shapecoder: a new method for visual quantification of body mass index in young children

Area of Science:

• Pediatric health
• Biometrics
• Obesity research

Background:

• Quantifying body mass index (BMI) visually in children is challenging due to limited tools.
• Accurate BMI assessment is crucial for monitoring child growth and identifying weight-related health risks.

Purpose of the Study:

• To evaluate the inter-rater reliability and validity of a novel 3D visual rating system for quantifying pediatric body mass index (BMI).
• To assess the system's sensitivity and specificity in identifying overweight/obesity and obesity in young children.

Main Methods:

• A 3D computer-based figure rating system (shapecoder) was applied to videotaped images of 242 children (mean age 5.9 years).
• Weight and height were measured, and shapecoder-derived BMI was compared against measured BMI.
• Inter-rater reliability, correlation, sensitivity, specificity, and predictive values were calculated.

Main Results:

• The shapecoder system demonstrated excellent inter-rater reliability (intraclass correlation coefficient = 0.98).
• Correlation between shapecoder-generated BMI and measured BMI was high (0.89).
• For overweight/obesity, sensitivity was 62% and specificity was 97%; for obesity, sensitivity was 65% and specificity was 99%.

Conclusions:

• The shapecoder system offers a reliable method for quantifying child BMI from video images.
• It provides fair sensitivity and good specificity for identifying overweight/obesity and obesity, representing an improvement over 2D methods.