Estimation of the body segment inertial parameters for the rigid body biomechanical models used in motion analysis

Abstract

Body segment inertial parameters (BSIPs) of the human body are key parameters in biomechanics to study the dynamics of human motion. BSIPs can be obtained in different ways including direct measurements on cadavers or photogrammetry and medical imaging on living humans, but they are more generally estimated by regression equations (based on those measurements). This chapter overviews three widely used regression equations reported by Winter (2009), de Leva (1996a), and Dumas et al. (2007a). These regression equations are presented for the head with neck, thorax, abdomen, pelvis, and right upper arm, forearm, hand, thigh, shank, and foot segments. The segment endpoints and segment reference frames defined at the time of the BSIPs assessment and regression computation are reviewed so that the reader can consider how they match with the construction of the rigid body biomechanical models they would like to use for motion analysis. The segment definitions and regression equations that remain undefined or unavailable are indicated, and some assumptions are proposed to amend them, where found applicable. The computation of the segment mass, position of center of mass, moments, and products of inertia from these regression equations are fully detailed, including the modification of the designation of the segment axes and the transformation from right to left segments.