Automatic Construction of Biventricular Statistical Shape Models

Abstract

This paper presents a method for the generation of landmarks in three-dimensional shapes and the construction of the corresponding 3-D statistical shape models. The technique relies on establishing shape correspondances via a volumetric non-rigid registration technique using multi-resolution B-spline deformations. This approach presents some advantages over previously published methods: it can treat multiple-part structures, and it requires less restrictive assumptions on the structure’s topology. In this paper we address the problem of building a three-dimensional statistical shape model of the left and right ventricles of the heart from 3-D Magnetic Resonance (MR) images. This application demonstrates the robustness and accuracy of the method in the presence of large shape variability and multiple objects.KeywordsIterative Close PointLabel ImageActive Shape ModelAutomatic ConstructionTraining ShapeThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.