Allometric models of adult regional body lengths and circumferences to height: Insights from a three‐dimensional body image scanner

Abstract

Objectives: Recent reports on body regional mass scalings to height have advanced understanding differences in adult heights. These studies resulted in conjectures on how regional lengths and circumferences may scale to height. We provide evidence for these conjectures by analyzing a large sample of regional limb, trunk, chest, and head lengths and circumferences in a large sample of US Army basic training recruits. Methods: Participants consisted of 10 271 males and 2760 females ages 17 to 21 years old who reported for basic training at Fort Jackson, SC. Participants were imaged by a three-dimensional (3D) body scanner for uniform sizing which yielded 159 body measurements of total mass, lengths and circumferences at regional sites of arms, legs, trunk, chest, and head. The allometric model, Body Measure i = α i H β i was applied to derive scaling exponents which were applied to estimate regional mass scalings. Results: Body mass scaled to height with powers of 2.0 (mean β ± SE, 1.98 ± 0.04, 1.93 ± 0.06). Arm and leg lengths scaled to exponents larger than 1.0 and head height and circumferences at regional sites scaled to exponents smaller than 1.0. The leg, arm, and trunk mass scaling exponents were all above 2.0. Head mass scaled to powers smaller than 2.0. Conclusions: The 3D scanner allowed hundreds of anthropometric measurements to be obtained within seconds. The ensuing analysis revealed that greater height yielded disproportional increases in limb lengths, limb mass and trunk mass. These analyses provide evidence that could not be previously measured that further both biomechanical and metabolic conjectures.

Torso Shape Improves the Prediction of Body Fat Magnitude and Distribution

“…Allometric scaling was assumed with the relationship , where is the measure, is the participant’s height and are parameters specific to each measure. Parameters were taken from [ 31 ] for measures of the arms and legs and [ 32 ] for measures of the torso.…”

Total body and regional surface area: Quantification with low‐cost three‐dimensional optical imaging systems

“…Body surface area (SA) and related body volume are features of the human phenotype incorporated into multiple thermophysiology and evolutionary biology models (Allen, 1877; Bergmann, 1847; Kasabova & Holliday, 2015; Welles et al, 2018). Humans vary widely in body dimensions and physical proportions, including the relationship of SA to body volume, with large phenotypic differences present across race/ethnic groups (Heymsfield, 2016) and individuals within selected groups (Watts et al, 2020). Body surface area has traditionally been among the more difficult features to the human phenotype to accurately quantify (Geoghegan, 1953; Pierson, 1962; Rodahl, 1952).…”

“…. Humans vary widely in body dimensions and physical proportions, including the relationship of SA to body volume, with large phenotypic differences present across race/ethnic groups (Heymsfield, 2016) and individuals within selected groups (Watts et al, 2020). Body surface area has traditionally been among the more difficult features to the human phenotype to accurately quantify (Geoghegan, 1953;Pierson, 1962;Rodahl, 1952).…”

Assessing human morphology using statistical shape analysis

“…Finally, these height normalised torso size measures were converted into z-scores, providing a common scale in units of standard deviations to enable subsequent stages of analysis. However, it has previously been shown that body proportions, such as segment lengths and girths, increase disproportionally with increased body height (244). Thus, to understand how each extracted body size measure would truly be expected to change with increased height, linear models were created using ordinary least squares regression to estimate the scaling exponent power, 𝛽 𝑖 , in the allometric models:…”

“…This method of normalisation adjusts measures for body height and assumes that the size of different parts of the body scale linearly with height. However, this was an inappropriate method of height normalisation, since it has been shown previously that regional body lengths and girths do not all scale linearly with body height, but in fact scale disproportionally with increased height (244). In future studies, scaling exponents should be determined for body size measures across all regions of the body, rather 124 than simply dividing my height to achieve more precise normalisation of body size, as demonstrated in Section 6.2.3.…”