Premium
Rapid‐accurate anthropometric body shape assessment with low‐cost novel 3D imaging system (391.2)
Author(s) -
Zhang Kang,
Zheng Jolene,
Gao Chenfei,
Thomas Diana,
Li Xin,
Heymsfield Steven
Publication year - 2014
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.28.1_supplement.391.2
Subject(s) - anthropometry , artificial intelligence , circumference , software , nuclear medicine , digital camera , monochrome , digital radiography , computer science , computer vision , mathematics , biomedical engineering , medicine , radiography , geometry , surgery , programming language
Anthropometric body dimensions are increasingly recognized as phenotypic markers of obesity risk. Available automated laser measurement systems provide accurate anthropometric dimensions but are costly. The study objective was to develop a low‐cost system that rapidly acquires digital morphology data that can be processed into anthropometric estimates. The core 3D system (KX16; TC2, Cary, NC) includes 16 Kinect (infrared laser projector and monochrome CMOS camera sensor) scanners positioned around the subject; we developed software that matches and composes raw scans and repairs missing/occluded regions to reconstruct complete 3D body geometry. Scan time is ~10 secs and image reconstruction‐digital measurements require 6‐8 min. The 3D system was evaluated in 36 subjects, age 20‐68 yrs and BMI 18.5‐41.4 kg/m 2 . Digital circumference measurements (mid‐arm; fore‐arm; thigh; ankle) were compared with corresponding tape measurements. The 3D‐ and tape‐measured respective circumferences were highly correlated (R2, 0.89, 0.69, 0.90, and 0.93; all p<0.01) with similar mean (±SD) results (tape/3D: 29.8±4.7 vs. 29.2±4.6; 18.5±2.3 vs. 18.5±2.5; 53.4±6.6 vs. 52.6±6.9; 26.1±4.1 vs. 26.9±4.0 cm) and non‐significant Bland‐Altman plots. Body shape and geometry can thus be quantified with acceptable accuracy using a relatively low‐cost digital imaging system coupled with novel image reconstruction software.