Study On Medical MR Image Segmentation Based On Fuzzy C - Means Clustering

In recent years, medical image segmentation is playing an increasingly important role in the field of clinical analysis and diagnosis, as an important part of medical image segmentation, MR image segmentation has attracted more and more attention. Since fuzzy C means clustering algorithm(FCM) based on the theory of fuzzy math can effectively describe the characteristics of MR image with high noise, low-contrast, fuzzy boundaries between different regions, so it has been widely used and applied. This paper makes FCM algorithm as research focus and will analyze and research fuzzy C means clustering theory and its application in medical MR image segmentation.The main work of this paper are as follows:(1) Analyzing and discussing the basic theory of FCM algorithm, researching the advantages and disadvantages of FCM algorithm for medical MR image segmentation, making analysis and comparison between different improved FCM algorithms with the research objects of brain MR image and the breast MR image.(2) Aiming at the drawbacks existing in the FCM algorithm of the slow speed of operation, result vulnerable to the initial value and the difficulty to deal with the inherent Rician noise of MR image, this paper presents a fast FCM algorithm combined with non-local means. The core of the algorithm are as follows:Firstly, using the non-local means algorithm to deal with the Rician noise, eliminating the impact of noise on segmentation result. Secondly, getting the initial cluster centers automatically according to the proposed rules of initial centers.Finally, the cluster centers should be as the initial cluster centers of fast FCM for the segmentation of the denoised image to solve the slow search speed and the problem that is easy to fall into local minima caused by the random selection of the initial clusters. Experimental results show that the proposed algorithm can quickly and efficiently segment the image, and is more robust to noise.(3) This paper presents a method of medical MR image segmentation which is FCM algorithm combined with level set. FCM is a method based on gray information and can effectively analyze the image gray imformation, but it lacks smoothness constraint, so it could not get the smooth segmentation boundary and segmentation range. Taking into account the advantages of level set for image segmentation, this paper presents a method combining level set model with FCM algorithm, first using FCM algorithm to segment the image to remove the noise and make gray difference of each part become larger, then using membership function obtained as initial condition and control parameter of level set to overcome the drawback of level set which depends on initial condition, improving the speed of curve evolution largely and geting a more refined result. Experimental results show this method is a better method of image segmentation.