Shape-based Tracking and Analysis of the Aorta in Cardiac MR Images Using Geometrically Deformable Templates

We have developed a new approach to shape-based tracking and analysis of cardiac MR images using geometrically deformable templates (GDTs). We propose to use an energyminimizing deformable template which can deform into similar shapes under the in uence of external forces. The degree of deformation of the template from its equilibrium shape is expressed by the amount of bending energy which is necessary to bend an idealized thinplate of metal. By minimizing this shape energy along with the image energy terms of the classic deformable model, the deformable template is attracted to objects in the image whose shape is similar to its equilibrium shape. The energy minimization problem of the deformable template is formulated in a probabilistic framework and solved by a stochastic relaxation technique, Simulated Annealing. We have applied the algorithm for the shapebased tracking and analysis of the aorta in MR cine sequences as well as for the segmentation and tracking of the myocardium.