Modelling of human torso shape variation inferred by geometric morphometrics

Area of Science:

• Anthropometry
• Biomedical Engineering
• Human Morphology

Background:

• Traditional body measurements offer limited insight into complex human body shapes.
• Three-dimensional (3D) imaging captures detailed surface data, presenting potential for advanced anthropometry in health.
• Existing 3D imaging studies analyze shape using traditional measures, not capturing full morphological complexity.

Purpose of the Study:

• To evaluate if geometric morphometrics (GM) shape measures provide additional insights into human morphology and mass distribution compared to traditional methods.
• To assess the explanatory power of current techniques in human torso shape variation using 3D imaging data.

Main Methods:

• Extracted scale-invariant features of torso shape from 3D imaging data of 9,209 participants (LIFE-Adult study).
• Utilized geometric morphometrics (GM) for statistical analysis of biological shape variation, applying the Procrustes paradigm.
• Employed partial least squares regression (PLSR) to model torso shape variations against traditional body measures.

Main Results:

• Traditional body measures explained 49.92% (males) and 47.46% (females) of torso shape variation.
• A significant portion of human morphological variation remains unexplained by current traditional measurement techniques.
• GM identified shape measures that offer complementary information beyond traditional anthropometry.

Conclusions:

• Geometric morphometrics (GM) provides a more comprehensive understanding of human body shape and mass distribution.
• GM-derived measures offer valuable, complementary data for assessing human morphology.
• Future research will explore the clinical epidemiology and health risk assessment utility of GM-based measures.