Landmark and Intensity-Based Images Elastic Registration Using Hierarchical B-Splines

Abstract

We present an elastic registration algorithm for the alignment of biological images. Our method combines and extends some of best techniques available in the context of medical imaging. We express the deformation field as a hierarchical B-splines model basing on the information of landmarks. The hierarchical B-splines model allows us to deal with a rich variety of deformations and match two images at increasing levels of detail. We solve the registration problem by minimizing the mutual information between the reference image and the test image. Results using MRI brain data are presented that the degree of matching is higher and the cost of time is less, compared with the algorithm which has not used the concept of hierarchical B-splines.