A Methodology for Constructing Geometric Priors and Likelihoods for Deformable Shape Models

Deformable shape models require correspondence across the training population in order to generate a statistical model for use as a future geometric prior. Traditional methods use fixed sampling and assume correspondence, or attempt to induce correspondence by minimizing variance. In this paper, we define a training methodology for sampled medial deformable shape models (m-reps) which generates correspondence implicitly via a geometric prior. We present quantitative results of the method applied to real medical images.