Validity of a 3-dimensional body scanner: comparison against a 4-compartment model and dual energy X-ray absorptiometry.

Abstract

Three-dimensional (3D) body scanner technology for body composition assessment is expanding. The aim of this study was to assess the validity of a 3D body scanner. One hundred and ninety-four participants (43% male; age: 23.52± 5.47years; body mass index: 23.98± 3.24kg·m-2) were measured using 3D scanner and a 4-compartment (4C) model utilizing dual-energy X-ray absorptiometry (DXA), air displacement plethysmography, and bioelectrical impedance spectroscopy. Dependent ttests, validity statistics including total error (TE), standard error of the estimate, constant error, and Bland-Altman analyses were utilized. Compared with 4C, 3D scanner fat mass (FM) [mean difference (MD; 3D-4C): 2.66kg ± 3.32kg] and percent body fat (%BF) (MD: 4.13%± 5.36%) were significantly (p< 0.001) over-predicted; fat free mass (FFM) was significantly underpredicted (MD: -3.15kg ± 4.75kg; p< 0.001). 3D demonstrated poor validity indicated by TE (%BF: 5.61%; FM: 4.50kg; FFM: 5.69kg). In contrast, there were no significant differences between 3D and DXA measures; 3D scanner demonstrated acceptable measurement for %BF (TE: 4.25%), FM (TE: 2.92kg), and lean mass (TE: 3.86kg). Compared with the 4C criterion, high TE values indicated 3D estimates were not valid. In contrast, 3D estimates produced acceptable measurement agreement when compared with DXA; an average overestimation of %BF by 5.31% (vs. 4C) and 4.20% (vs. DXA) may be expected. Novelty: 3D body composition estimates are not valid compared with the 4-C criterion model. 3D estimates appeared to be more valid in females, compared with males. When compared with DXA, 3D estimates were acceptable.