Image registration is the task of aligning two or more images into the same reference frame to compare or distinguish the images. The majority of registration methods deal with registering only two images at a time. Recently, a clustering method that concurrently registers more than two multi-sensor images was proposed, dubbed ensemble clustering. In this thesis, we apply the ensemble clustering method to deformable registration scenario for the first time. Non-rigid deformation is implemented by a FFD model based on B-splines. A regularization term is added to the cost function of the method to limit the topology and degree of the allowable deformations. However, the increased degrees of freedom in the transformations caused the Newton-type optimization process to become ill-conditioned. This made the registration process unstable. We solved this problem by using the matrix approximation afforded by the singular value decomposition (SVD). Experiments showed that the method is successfully applied to non-rigid multi-sensor ensembles and overall yields better registration results than methods that register only 2 images at a time. In addition, we parallelized the ensemble clustering method to accelerate the performance of the method. The parallelization was implemented on GPUs using CUDA (Compute Unified Device Architecture) programming model. The GPU implementation greatly reduced the running time of the method.

Image registration is the task of aligning two or more images into the same reference frame to compare or distinguish the images. The majority of registration methods deal with registering only two images at a time. Recently, a clustering method that concurrently registers more than two multi-sensor images was proposed, dubbed ensemble clustering. In this thesis, we apply the ensemble clustering method to deformable registration scenario for the first time. Non-rigid deformation is implemented by a FFD model based on B-splines. A regularization term is added to the cost function of the method to limit the topology and degree of the allowable deformations. However, the increased degrees of freedom in the transformations caused the Newton-type optimization process to become ill-conditioned. This made the registration process unstable. We solved this problem by using the matrix approximation afforded by the singular value decomposition (SVD). Experiments showed that the method is successfully applied to non-rigid multi-sensor ensembles and overall yields better registration results than methods that register only 2 images at a time. In addition, we parallelized the ensemble clustering method to accelerate the performance of the method. The parallelization was implemented on GPUs using CUDA (Compute Unified Device Architecture) programming model. The GPU implementation greatly reduced the running time of the method.

Registration of a group of images generally only gives a pointwise, dense correspondence defined over the whole image plane or volume, without having any specific description of any common structure that exists in every image. Furthermore, identifying tissue classes and structures that are significant across the group is often required for analysis, as well as the correspondence. The overall aim is instead to perform registration, segmentation, and modelling simultaneously, so that the registration can assist the segmentation, and vice versa. However, structural information does play a role in conventional registration, in that if the registration is successful, it would be expected structures to be aligned to some extent. Hence, we perform initial experiments to investigate whether there is explicit structural information present in the shape of the registration objective function about the optimum. We perturbed one image locally with a diffeomorphism, and found interesting structure in the shape of the quality of fit function. Then, we proceed to add explicit structural information into registration framework, using various types of structural information derived from the original intensity images. For the case of MR brain images, we augment each intensity image with its own set of tissue fraction images, plus intensity gradient images, which form an image ensemble for each example. Then, we perform groupwise registration by using these ensembles of images. We apply the method to four different real-world datasets, for which ground-truth annotation is available. It is shown that the method can give a greater than 25% improvement on the three difficult datasets, when compared to using intensity-based registration alone. On the easier dataset, it improves upon intensity-based registration, and achieves results comparable with the previous method.