Multi-Modality Medical Image Registration Basedon Maximization of Mutual Information
基于最大互信息的多模医学图象配准

In this paper a maximization of mutual information based multi -modality medical image registration method is described. The method presented in this paper applies mutual information to measure the information redundancy b etween the intensities of corresponding voxels in both images, which is assumed to be maximal if the images are geometrically aligned. MI is used as a measure o f similarity of two images. There exist many important technical issues to be so lved about the method such as how to compute MI more accurately and how to obtai n the maximization of MI, which are seldom mentioned in published papers. In thi s paper we provide some implementation issues, for example, subsampling, PV inte rpolation, outlier strategy. Powell searching algorithm is used which does not c ompute gradients. The combination of these computation techniques and searching strategy leads to a fast and accurate multi-modality image registration. The re gistration results of 3D human brain volume data of 41 CT-MR and 35 PET-MR fro m seven patients are validated to be subvoxel. The registration method is promis ing in clinical use.