Multi-resolution CT-MR brain image registration using normalized mutual information

Medical image registration has been applied to the diagnosis of cardiac studies and a variety neurological disorders including brain tumors. This paper presents a wavelet-based multi-resolution implementation of Computed Tomography (CT) and Magnetic Resonance (MR) brain image registrations based on normalized mutual information. The implementation gives improved accuracy with no loss of speed This is because wavelet-based multi-resolution preserves most of the important features of the original data, even at a low resolution, so accurate subvoxel coregistration can be achieved In this paper we illustrate this implementation by presenting the results of tests on the data of 7 patients (41 pairs of CT-MR). Registration accuracy is evaluated by J.M. Fitzpatrick at Vanderbilt University, as part of the Retrospective Registration Evaluation Project.