A Dynamic Surface Model for Segmentation of Cardiac MRI Images and its Usage in Cardiac Wall Motion Tracking

摘要

This paper presents the application of a new active contour algorithm for object detection in 3D space. The introduced model is based on techniques of curve evolution for segmentation and level sets. We minimize an energy which can be considered as a particular case of the minimal partition problem. In the level set formulation, the problem becomes a "mean-curvature flow"-like evolving the active contour, which will stop on the desired boundary. However, the stopping term does not depend on the gradient of the image, as in the classical active contour models, but is instead related to a particular segmentation of the image. The method can be put into a 3D level-set framework using a Lipschitz function phi for automatic topology changes. The novel developed 3D algorithm was used for 3D segmentation of endocardial wall in the left ventricle. The experimental results demonstrated the efficiency of the proposed scheme. The segmented cardiac image was then used by our previously developed 3D tracking algorithm to track the endocardial wall motion which resulted in a fully automatic 3D point-wise tracking software for analyzing 3D cardiac MRI images.