Generalized robust point matching using an extended free-form deformation model: application to cardiac images

In this paper we present a novel and automated technique based on the generalized robust point matching (G-RPM) framework using an extended free-form deformation (EFFD) model for analysis of left ventricular (LV) motion. Unlike free-form deformations (FFDs) that employ parallelepipedical lattice shapes, the EFFD models use arbitrarily-shaped lattices. Our technique extends the set of possible deformations, and thus a more accurate estimate of LV deformation can be obtained. The computation is efficient since the number of control points is largely reduced. In particular for cardiac motion tracking, it requires roughly segmenting the myocardial boundary in only the first time frame and no more segmentation is needed. We evaluate the approach on sequences of in-vivo cardiac magnetic resonance images.