Research On Automatic Medical Image Segmentation Algorithm Based On Fuzzy Set And Level Set

Medical image segmentation is one of the earliest application fields of image segmentation, and it also is the key point of medical image analysis. In addition, medical image segmentation, it is important and crucial in application of clinical medicine, is playing an increasingly important role in medical imaging. The definition of medical image segmentation is to separate object regions of interest such as tumors, cerebral contusions, diseased tissues, hematomas, cells, diffuse brain swellings and blood vessels from image background through the advanced computer technique. It is not only the basis of three-dimensional reconstruction and visualization of medical image, but also the premise of quantitatively measuring the boundary, shape and section area of diseased tissues. In short, medical image segmentation is of great significance in image processing and its applications.In recent years, in order to obtain some automatic medical image segmentation algorithms, it has some characteristics of high accuracy, real-time and simple operability, a development tendency of methods for medical image segmentation is to merge various existing methods while exploring new methods and new ideas. Currently, some existing fuzzy level set segmentation algorithms are a simple combination of two methods, which do not realize the automation of medical image segmentation and need to regulate manually control parameters of level set method. So this paper researches on fuzzy set theory and level set method, the main work is summarized as follows:(1) We research methods for medical image segmentation based on fuzzy set theory and propose an improved FCM algorithm.In this part, on the basis of analyzing fuzzy set theory, we key research FCM algorithm. According to the poor anti-noise performance of traditional FCM algorithm, this paper proposes an improved FCM algorithm. Experimental results show that, compare with traditional FCM algorithm, the improved FCM algorithm has some merits of powerful antinoise ability and high precision. (2) We research methods for medical image segmentation based on level set and present a new fuzzy level set algorithm.In this part, on the basis of theoretical analysis on curve evolution theory and mathematical model of level set, we key research evolution equations of level set method when it is used to segment medical image. Moreover, we also study the optimization criterion, action and significance of control parameters. However, aiming at disadvantages of traditional level set method, we also key research LCM model. Besides that, in this paper, we present a new fuzzy level set algorithm by combining the improved FCM algorithm and LCM model. Experimental results show that the new algorithm in this paper has some advantages such as fast segmentation speed, easy operation and strong noise immunity. Moreover, it makes medical image segmentation automation.(3) Computer-assisted medical imaging segmentation system is designed and developed.In order to compare differences qualitatively and quantitatively between new methods and other methods on the same platform, I also develop a computer-assisted medical imaging segmentation system. In this system, we can not only use a variety of methods to segment the same medical image but also compare performance, accuracy and running speed of these methods in quantitative and qualitative way.