Mutual Information Based Medical Image Registration Algorithm

Medical image registration, developed in 1990's, is one of the important research on medical image processing. Image fusion is based on medical image registration. It overcomes the limit of only using one image, effectively brings together various information about the two images to show them on an image. Thus, the information, including human anatomization, human physiology and human pathology, can all be provided to make for the clinical diagnosis and the treatment. In the paper, medical image registration and the related technology is studied. And the principle and detailed steps of medical image registration is described. Then the paper discuss the main methods of medical image registration, and emphasize on the medical image registration based on mutual information, finally the algorithm design and realization of it are achieved.Firstly, it is described in the paper that the background, development actuality and clinical practices of medical image registration. Then the paper describes the principle of medical image registration. And the detailed steps how to finish the registration also are analyzed, in which the geometry transform, the interpolation method, the similarity measure, the optimized algorithm and their effects on registration are thoroughly discussed. Meanwhile, both the classification of the registration method and the concerned evaluation are surveyed. Next, the paper studies the medical image registration based on mutual information, the background and the foundation concept of it are introduced, and the difficulties of this method is carried on the discussion, so as its research emphasis. On the base of the theory, the following part introduces the similarity measure of generalized mutual information, expatiate the medical image registration using the measure. Finally, the registration experiments of mutual information are done under Matlab7.0 environment, in which the three-dimensional medical images of the human brain are taken as the testing data. The simulation achieves the medical image registration based on mutual information, which results confirm the accuracy and the robust of the registration algorithm, and the registration of other two similarity measure are also simulated, which results are compared with that of mutual information, then some advantage and shortcomingexisting in the method of mutual information, is summarized by the compare with other registration methods.