Registration of Multimodality Medical Imaging of Brain using Particle Swarm Optimization

In present work we have introduced nonlinear affine registration method to incorporate the anatomic body deformation. The present technique has been developed for registration of section of human brain using CT and MR modalities. Present study related to image registration of different modality imaging is based on 2-D/2-D affine registration technique. Automatic registration has been achieved by maximization of a similarity measure and which is the correlation function of two images. The proposed method has been implemented by choosing a realistic, practical transformation and optimization techniques. Since similarity metric is a non-convex function and contains many local optima, choice of search strategy for optimization is important in registration problem. There exist many optimization schemes, most of which are local and require a starting point. Presently, we have implemented multiresolution based particle swarm optimization technique to overcome this problem.