Validation of a three-dimensional body scanner for body composition measures.

The accuracy of an infrared three-dimensional (3D) body scanner in determining body composition was compared against hydrostatic weighing (HW), bioelectrical impedance analysis (BIA), and anthropometry. A total of 265 adults (119 males; age = 22.1 ± 2.5 years; body mass index = 24.5 ± 3.9 kg/m2) had their body fat percent (BF%) estimated from 3D scanning, HW, BIA, skinfolds, and girths. A repeated measures analysis of variance (ANOVA) indicated significant differences among methods (p < 0.001). Multivariate ANOVA indicated a significant main effect of sex and method (p < 0.001), with a non-significant interaction (p = 0.101). Bonferroni post-hoc comparisons identified that BF% from 3D scanning (18.1 ± 7.8%) was significantly less than HW (22.8 ± 8.5%, p < 0.001), BIA (20.1 ± 9.1%, p < 0.001), skinfolds (19.7 ± 9.7%, p < 0.001), and girths (21.2 ± 10.4%, p < 0.001). The 3D scanner decreased in precision with increasing adiposity, potentially resulting from inconsistences in the 3D scanner's analysis algorithm. A correction factor within the algorithm is required before infrared 3D scanning can be considered valid in measuring BF%.