Validity and reliability of total body volume and relative body fat mass from a 3-dimensional photonic body surface scanner

Abstract

Objective Three-dimensional photonic body surface scanners (3DPS) feature a tool to estimate total body volume (BV) from 3D images of the human body, from which the relative body fat mass (%BF) can be calculated. However, information on validity and reliability of these measurements for application in epidemiological studies is limited. Methods Validity was assessed among 32 participants (men, 50%) aged 20–58 years. BV and %BF were assessed using a 3DPS (VitusSmart XXL) and air displacement plethysmography (ADP) with a BOD POD® device using equations by Siri and Brozek. Three scans were obtained per participant (standard, relaxed, exhaled scan). Validity was evaluated based on the agreement of 3DPS with ADP using Bland Altman plots, correlation analysis and Wilcoxon signed ranks test for paired samples. Reliability was investigated in a separate sample of 18 participants (men, 67%) aged 25–66 years using intraclass correlation coefficients (ICC) based on two repeated 3DPS measurements four weeks apart. Results Mean BV and %BF were higher using 3DPS compared to ADP, (3DPS-ADP BV difference 1.1 ± 0.9 L, p<0.01; %BF difference 7.0 ± 5.6, p<0.01), yet the disagreement was not associated with gender, age or body mass index (BMI). Reliability was excellent for 3DPS BV (ICC, 0.998) and good for 3DPS %BF (ICC, 0.982). Results were similar for the standard scan and the relaxed scan but somewhat weaker for the exhaled scan. Conclusions Although BV and %BF are higher than ADP measurements, our data indicate good validity and reliability for an application of 3DPS in epidemiological studies.