Accuracy and Precision of 3D Optical Imaging for Body Composition and their Associations to Metabolic Markers by Age, BMI, and Ethnicity

Abstract

Background: Recent three-dimensional optical (3DO) imaging advancements have provided a more accessible, affordable, and self–operating opportunities for assessing body composition. 3DO is accurate and precise compared to clinical measures measured by dual–energy X–ray absorptiometry (DXA) in diverse study samples. However, the accuracy and precision of an overall 3DO body composition model in specific subgroups is unknown. Therefore, this study aimed to evaluate 3DOs accuracy and precision by subgroups of age, body mass index (BMI), and ethnicity/race. Methods: A cross–sectional analysis was performed using the Shape Up! Adults study. Each participant received duplicate 3DO and DXA scans. 3DO meshes were digitally registered and reposed using Meshcapade to standardize the vertices and pose. Principal component analysis was performed on the registered 3DO meshes to orthogonalize and reduce the dimensionality of the data. The resulting principal components estimated DXA whole–body and regional body composition using stepwise forward linear regression with five–fold cross–validation. Duplicate 3DO and DXA scans were used for test–retest precision. Student t–test was performed between 3DO and DXA by subgroup to determine significant differences. One–way ANOVA determined if intra–group precision had significant differences. P–value < 0.05 was considered statistically significant. Results: Six hundred thirty–four participants (females = 346) had completed the study at the time of the analysis. 3DO total fat mass (FM) and fat–free mass (FFM) in the entire sample achieved R2s of 0.94 and 0.92 with RMSEs of 2.91 kg and 2.76 kg, respectively, in females and similarly in males. 3DO total FM and FFM achieved a %CV (RMSE) of 1.76% (0.44 kg) and 1.97% (0.44 kg), while DXA had a %CV (RMSE) of 0.98% (0.24 kg) and 0.59% (0.27 kg), respectively, in females and similarly in males. There were no mean differences by age group (p–value > 0.068). However, there were mean differences for underweight females, NHOPI females and males, and Asian and black females (p–value < 0.038). There were no significant differences among the subgroups for precision (p–value > 0.109). Conclusion: A single 3DO body composition model derived from a highly–stratified dataset performed well against DXA with minimal differences detected for accuracy and precision. Adjustments to specific subgroups may be warranted to improve the accuracy in those that had significant differences. Nevertheless, 3DO produced accurate and precise body composition estimates that can be used on diverse populations.

Evaluation of Body Shape as a Human Body Composition Assessment in Isolated Conditions and Remote Environments

“…3DO can identify signals associated with muscle, bone, and frailty 13 . 3DO scanners use depth cameras to scan the surface of the body to provide accurate and precise automated anthropometry and a high-resolution 3D mesh 14,15 , which is used to create statistical shape models to predict body composition 13,[16][17][18][19] . A recent paper showed that reposing these meshes to a standardized pose before analysis improved the accuracy and precision regardless of initial pose 20 .…”

Evaluation of body shape as a human body composition assessment in isolated conditions and remote environments

“…3DO can identify signals associated with muscle, bone, and frailty 13 . 3DO scanners use depth cameras to scan the surface of the body to provide accurate and precise automated anthropometry and a high-resolution 3D mesh 14,15 , which is used to create statistical shape models to predict body composition 13,[16][17][18][19] . A recent paper showed that reposing these meshes to a standardized pose before analysis improved the accuracy and precision regardless of initial pose 20 .…”