A deformable image-based registration approach to obtain shape correspondence for statistical shape modeling of finger bones

Annkristin Lange, Stefan Heldmann, Jan Hendrik Moltz, Lars Walczak, Christiane Engel, Mia Detering, Anna Rörich, Felix Güttler, Sinef Yarar-Schlickewei, Joachim Georgii

Abstract: In this paper we introduce a registration approach based on deformable images to obtain the necessary shape correspondences to compute statistical shape models. We focus on the modelling of finger bones and have therefore developed an initialisation strategy based on cylindrical coordinates. First results of shape models for single bones and whole fingers are presented.