CT-MR image registration in liver treatment by maximization of mutual information

CT and MR images have been used in the liver disease treatment. In order to merge the different information into one image, image registration and data fusion must be adopted. In this paper, image registration and data fusion for MR and CT images of liver was studied. Firstly, the maximization of mutual information method was proposed to register the images. Then, some optimization algorithms, such as Powell, Brent, Golden Section and the second-degree interpolation algorithms, were applied. Finally, the post-registration images were fused with a wavelets-based scheme and the different information was merged into a new image. The experiment results demonstrated the effectiveness of the fusion scheme.