WSEAS Transactions on Signal Processing
Print ISSN: 1790-5052, E-ISSN: 2224-3488
Volume 12, 2016
Locally Adaptive Topology Preservation for Diffeomorphic Registration in Medical Imaging
Authors: , , , ,
Abstract: Diffeomorphic registration has become an active research field presently in medical image registration because of its the differential transformation with invertibility between anatomic individuals. In this paper, we propose a novel method named Locally Adaptive Topology Preservation for Diffeomorphic Registration, which is able to obtain accurate approximation for the local tangent space on the Lie group manifold and yield more plausible diffeomorphisms for spatial transformations. In order to incarnate the local geometric structure of the Lie group, the local linear approximation is adaptively optimized by selecting appropriate neighborhoods for each sample point. Furthermore, we investigate the Lie group structure of the Symmetric Positive Definite (SPD) matrices and evaluate the effectiveness of the algorithm by utilizing several sets of brain images. Experimental results demonstrate that our algorithm has a higher degree of topology preservation on a dense high-dimensional deformation field and performs better in the noisy setting.
Search Articles
Keywords: Diffeomorphic registration, Lie group, Neighborhood selection, Symmetric positive definite matrices, Medical images
Pages: 28-38
WSEAS Transactions on Signal Processing, ISSN / E-ISSN: 1790-5052 / 2224-3488, Volume 12, 2016, Art. #4