Understanding the motion of the heart through the cardiac cycle can give useful insight for a range of different pathologies. In particular, quantifying regional cardiac motion can help clinicians to better determine cardiac function by identifying regions of thickened, ischemic or infarcted tissue. In this work we propose a method for cardiac motion analysis to track the deformation of the left ventricle at a regional level. This method estimates the affine motion of distinct regions of the myocardium using a near incompressible non-rigid registration algorithm based on the Demon's optical flow approach. The global motion over the ventricle is computed by a smooth fusion of the deformation in each segment using an anatomically aware poly-affine model for the heart. We apply the proposed method to a data-set of 10 volunteers. The results indicate that we are able to extract reasonably realistic deformation fields parametrised by a significantly reduced number of parameters compared to voxel-wise methods, which better enables for statistical analyses of the motion. © 2013 Springer-Verlag.
CITATION STYLE
McLeod, K., Seiler, C., Sermesant, M., & Pennec, X. (2013). A near-incompressible poly-affine motion model for cardiac function analysis. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 7746 LNCS, pp. 288–297). https://doi.org/10.1007/978-3-642-36961-2_33
Mendeley helps you to discover research relevant for your work.