Premium
3D Imaging Technology for Human Body Size and Shape Analysis: Comparison of Three Different Acquisition Technologies
Author(s) -
Bourgeois Brianna,
Latimer Dustin,
Stannard Casey R,
Romeo Laurel,
Li Xin,
Shepherd John,
Ng Bennett K,
Heymsfield Steven B
Publication year - 2017
Publication title -
the faseb journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.709
H-Index - 277
eISSN - 1530-6860
pISSN - 0892-6638
DOI - 10.1096/fasebj.31.1_supplement.433.1
Subject(s) - anthropometry , trunk , nuclear medicine , medicine , biomedical engineering , ecology , biology
Anthropometric measurements of body size and shape are widely used in diagnosing, monitoring, and assessing the risk of developing adiposity‐related chronic metabolic diseases. Recent technological advances have extended anthropometry beyond traditional tape‐derived measurements to automated 3D digital image acquisition devices that can rapidly acquire hundreds of linear, circumferential, and volumetric body dimensions. The aim of the current study was to compare measures across three imaging devices differing in acquisition and data processing technology (16 stationary Microsoft Kinect Cameras (MKC)[KX16, TC 2 ]; and single vertical movable optical light‐coding camera [Proscanner, Fit3D Inc.] or stationary MKC V2 camera [Styku, Inc.] with 360 0 rotating platforms) and against traditional reference methods including total body volume (TBV) by air displacement plethysmography (ADP), circumferences by tape anthropometry, and regional body volume by dual‐energy x‐ray absorptiometry (DXA). Subjects were 32 healthy adults (15 male, 17 female). TBVs measured by all three systems were highly inter‐correlated and significantly correlated with ADP‐derived TBV (respective R 2 's, 0.94, 0.97, and 0.96; all p<0.001). Regression line slopes were non‐significant in Bland‐Altman plots for KX16 and Fit3D and significant (p=0.001) for Styku. Relative to tape measurements, good correlations were observed for all circumferences derived by 3D devices (e.g., right bicep, R 2 's, 0.67, 0.94, 0.86; hip, R 2 's 0.67, 0.74, 0.83). Regional body volumes measured by each system were highly correlated with corresponding DXA volumes (e.g., trunk, R 2 's, 0.93, 0.94, 0.96; all p<0.001). Potentially research and clinically useful 3D imaging systems compare favorably with corresponding reference methods with similar overall results largely independent of acquisition technology.