We present a novel approach to nonrigid registration of volumetric multimodal medical data. We propose a new regularized template matching scheme, where arbitrary similarity measures can be embedded and the regularization imposes spatial coherence taking into account the quality of the matching according to an estimation of the local structure. We propose to use an efficient variation of weighted least squares termed normalized convolution as a mathematically coherent framework for the whole approach. Results show that our method is fast as accurate.
Abstract. We propose a novel approach to landmark-based medical image registration based on the geostatical method of Kriging prediction. Our method exploits the spatial statistical relation between two images, as estimated using generalpurpose registration algorithms, in order to construct an optimum predictor of the displacement field. High accuracy is achieved by using an estimated spatial model of the displacement field directly from the image data, while practically circumventing the difficulties that prevented Kriging from being widely used in image registration.
Abstract-To a great extent, the success of advanced image-guided medical procedures hinges on non-rigid volume registration. For example, nonrigid registration must be applied in interventional approaches where intraoperative information is used to update high-quality preoperative data; in follow-up studies in order to assess time-evolution of development; aging, pathology or treatment; and in many other applications including intersubject variability and population-based atlas construction.In this paper we examine several computational schemes that warps one volumetric dataset onto another. We also explore the inevitable trade-off between the computational load and the incorporation of sophisticated similarity measures necessary for multimodal volumes. Estimated deformation fields are based on the variational formulation of Partial Derivative Equations (PDEs), which includes a similarity and a regularization term. We compare numerical solutions to this problem using the Euler-Lagrange equations (EL), the Finite Elements discretization (FE), and Decoupled Optimization over the possible deformations (DO).
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.