Validation of a new method for bone motion measurement by soft-tissue artifact compensation through spatial interpolation

Localizing a bone hidden behind soft tissue is crucial in biomechanics. A typical approach consists in placing markers on the skin, measuring their position/orientation, and from these measurements, estimating the position and orientation of the bone hidden behind the skin while compensating of soft tissue deformations. In this paper, we present a new method to address this problem. It requires to initially record scattered spacial transformations between the rigid body and the markers. Then, a natural neighbors interpolation algorithm, modified to apply to homogeneous transformations, is proposed. The presented method is validated on a robot manipulator on which soft tissue is placed.