Estimation of body segment parameters using dual energy absorptiometry and 3-D exterior geometry

Abstract

Body segment parameters (BSP) are essential inputs in the kinetic analysis of human motion. These data include mass, center of mass and moments of inertia, and are usually obtained from population-specific predictive equations. Several limitations remained with the use of cadaveric measurement. These include limited sample size, as well as discrepancies in age and morphology. The limitations motivated the development of direct BSP measurements on living subjects. However, other limitations such as exposure to radiation, high cost and sheer lack of facilities that are readily available evolved. In addition, some methods provide only twodimensional (2-D) measurements while others require intensive tomographic images for three-dimensional (3-D) reconstruction. This article proposes a novel technique for reconstructing subject-specific 3-D body mass distribution using from Dual Energy Absorptiometry X-ray imaging and 3-D exterior geometry. The accuracy of our method was assessed by calculating percent error between the criterion based on total body mass and our proposed total body mass measurements (TBM). We found less than one percent error between the criterion and our TBM measurements for subjects of four different body types: endormorph, mesomorph, ectomorph and central. We found significant differences between our proposed method as compared to cadaver-based and livingbased models for segment mass (SM) at the hand; center of mass (CM) at the thigh, shank, foot, head and neck, trunk, upper arm, forearm and hand; and moments of inertia at the hand, shank, foot and trunk. Extensions of the research could further examined the influence of other BSP measurements derived from our proposed method and the four selected estimation models, as well as their impact on the inverse dynamics solutions in human motion.