Document Type



The purpose of this study was to determine the optimal number of rigid body segments to sufficiently represent the trunk movements, using Akaike’s information criterion. The trunk in static and dynamic conditions was modelled with one, two, three, or six linked rigid-body representations. The difference in the three-dimensional position between the actual and modelled data was calculated to quantify how well these models describe the actual trunk kinematics. The Akaike’s information criterion was calculated using the difference in position data to evaluate the goodness-of-fit for each model. Our findings suggest that two-linked rigid-body representation may be good enough when analysing trunk movements except when the movement includes a large axial rotation, for which the three-linked rigid-bodies would be better. These results would be useful in determining the optimal number of rigid body representation to sufficiently represent the trunk movements.