New Algorithms Of 3D Medical Image Rigid Registration

Medical image registration is one of the important techniques in the field of medical image processing. It also takes a more and more important role in clinical diagnosis and therapy. Although lots of researches on medical image rigid registration have been made for many years, there are still some drawbacks within the current main methods, which have to be corrected further and better for the clinical applications.There are two main kinds of methods in medical image rigid registration, which are feature-based and statistic-based. In this paper, one of the main algorithms of these two methods was implemented, which were the contour-based registration and the mutual-information-maximization-based registration, respectively. The advantages and disadvantages of the algorithms were analyzed. On the basis of these analyses, we originally put forward two new algorithms. Firstly, we presented the notion of fuzzy edge field (FEF). Being the feature of images, FEF could be extracted automatically without any manual interaction and used to strengthen the similarity between images. At the same time, a new registration algorithm based on fuzzy edge similarity (FES) was presented. Secondly, due to the inappropriate selection of overlapped regions under the current framework in statistic-based methods, we presented a new framework, which was called as the framework with same overlapped regions (FW-SOR). Under this framework, some statistic-based algorithms were united as conditional-entropy (CE)-based algorithm.We obtained many medical images, which were the standard images for image registration. A lot of experiments were done on these image data and the evaluations of the results in our experiments were provided by the providers of the images. It proved that the robustness of these new algorithms was greatly improved compared with the traditional algorithms.