Background

Childhood obesity is common and work is ongoing to identify risk factors which could be targets for intervention. However, many large existing data sets which could be used to advance the science in obesity research do not include measured weight and height.

Estimating a child’s anthropometrics visually is very difficult. Even trained health care workers have relatively low accuracy and are only slightly better than parents. A number of studies have therefore sought to use figure rating scales to provide a visual metric for estimating body mass index.

Figure. Sample of a 2-D rating scale

Given the success and common use of this 2-dimensional (2-D) rating scale for adults, there have been subsequent attempts to generate such a 2-D rating scale for application to children. The accuracy of 2D scales has not been directly evaluated. Rather, the scales’ validity has been described by comparing children’s or mother’s identification of body type from the scale in comparison to actual measured body mass index. Given the potential bias inherent in children’s self-identification of body type or mother’s identification of their child’s body type, this approach confounds social idealization of body image with validation of the scale.

To address these issues, a new software tool based on 3D laser scans of child body shape has been developed. ShapeCoder guides a rater through a series of rotatable, 3D figures of standing children generated from a statistical body shape model (SBSM). At each of four levels, the rater chooses the ShapeCoder figure that best matches the body shape of the subject. The underlying SBSM associates each presented figure with a BMI.

How ShapeCoder WEB Works...

"UNDER REVIEW..."