Background: As obesity increases throughout the developed world, concern for the health of the population rises. Obesity increases the risk of metabolic syndrome, a cluster of conditions associated with type-2 diabetes. Correctly identifying individuals at risk from metabolic syndrome is vital to ensure interventions and treatments can be prescribed as soon as possible. Traditional anthropometrics have some success in this, particularly waist circumference. However, body size is limited when trying to account for a diverse range of ages, body types and ethnicities. We have assessed whether measures of torso shape (from 3D body scans) can improve the performance of models predicting the magnitude and distribution of body fat.
Methods: From 93 male participants (age 43.1 ± 7.4) we captured anthropometrics and torso shape using a 3D scanner, body fat volume using an air displacement plethysmography device (BODPOD®) and body fat distribution using bioelectric impedance analysis.
Results: Predictive models containing torso shape had an increased adjusted R2 and lower mean square error when predicting body fat magnitude and distribution.
Conclusions: Torso shape improves the performance of anthropometric predictive models, an important component of identifying metabolic syndrome risk. Future work must focus on fast, low-cost methods of capturing the shape of the body.
Keywords: 3D body scan; anthropometry; body shape; fat distribution; fat volume; metabolic syndrome; multiple linear regression.