Medical Image Fusion Algorithm

Medical imaging technology is a method which utilizes all kinds of imaging devices to observe the internal organization, visceral shape and functional chage of the human body. Different medical imaging devices offer different medical information, and play different roles in clinical diagnosis and therapy. Medical image fusion can synthesize the complementary information inside those two processes, gain an image with better effect including more complete information, and supply more accurate data for clinical diagnosis and therapy. This thesis explores and researches wavelets-based fusion method and intelligentized fusion method with the discussing of their theories and algorithms.Wavelets-based fusion method is a relative better method from mathematical theory to arithmetic model. In this part of research, a new fusion algorithm used to fuse anatomical images such as CT and MRI is put forward first. The point of this algorithm is an optimization of registration process to earn two images which are better registered with the improvement of cross correlation method. Then gain another fusion method applied in anatomical image and functional image by expanding and improving this new fusion. It aims at the notequal resolution of an anatomical image and a functional image taken from the same scene, fuses the functional image with the approximate information of the decomposed anatomical image which is totally different from the usual interpolation method. Inaccurate information can be avoided in this algorithm and the execution speed rises.Intelligentized fusion method is one of the methods that is at early stage of development, but has superperformance and great potential development. In the exploration to this kind of method, it focuses on the fusion method based on the fuzzy logic, applies one fuzzy algorithm on the basis of simple rules, separates backgrounds, outlines and details in images through the definition of membership functions, replaces the strict classification with fuzzy clustering, simulates by using the Matlab fuzzy-logic tool box, and gets the best fusion image through the modification of membership function and rules. Then attain one neural-fuzzy method by connecting to the neural network, uses both the learning capability of neural network and thedecision-making capability of fuzzy logic. The training of membership functions is based on least square method and back propagation associated with gradient descent method. Finally the analysis and simulation are carried through.At the end of this thesis, it analyzes and compares these two fusion methods, and summarizes the advantages and disadvantages respectively. It also predicts the future trend of development and direction of medical image fusion technology.