Research On Medical Image Segmentation Algorithm Based On Level Set Method

Medical image segmentation takes a crucial part in identification and diagnosis oflesions. It has practical significance for biomedical field. However, due to the individualdifferences, and other uncertain effects on medical images, accurate segmentation is avery difficult topic in Processing and Analysis of Medical Images.Recently, level set method has become a research hotspot in the field of imagesegmentation and achieved a good performance while addressing the imagesegmentation problem．Compared with the traditional image segmentation methods, theimage segmentation method that based on level set has shown great advantage in thesegmentation of the complex medical images.This paper mainly study image segmentation of active contour model based onlevel set. First, the theory basis and numerical calculation of level set were introduced,and then we focus on two different GAC models in this paper. Li Chunming modelwhich is based on gradient information and C-V model based on region information.Based on the drawbacks and advantages of these two models, we introducepenalty-function to C-V mode, and proposed a C-V model that without re-initialization.Finally, a new fuzzy level set algorithm has been proposed for medical imagesegmentation. It utilizes fuzzy clustering as the initial level set function. Compared withthe traditional FCM algorithms, the enhanced FCM algorithms with spatial informationcan approximate the boundaries of interest well and the level set equation is modifiedwith variable balloon forces. In addition, the new algorithm estimates the controllingparameters from fuzzy clustering automatically. This has reduced manual intervention.At the end of the article performance evaluation has been carried out with medicalimages. The results show that the algorithm was effective.