期刊名称：International Journal of Hybrid Information Technology
印刷版ISSN：1738-9968
出版年度：2016
卷号：9
期号：11
页码：199-206
出版社：SERSC
摘要：Prostate segmentation from MRI is a necessary first step and plays a key role in different stages of clinical decision making process.In this paper,we propose aMR T1 Image segmentation method of the prostate based on distance regularized level set evolutionwith a deformableshape prior.To smooth the prostateimage to reduce thenoise,we introducea prostate image with Gaussian kernel and get an edge indicator. To avoid the leakageinduced on account of prostate boundaries missing, orblendingwith surrounding tissues, we introducepriori shape informationtoconstruct an energy function with a distance regularizationterm and an external shape energyterm containing the edge indicator and minimize it by solving the gradient flow which can be implemented with a finitedifference scheme.To verify the MRI segmentation method of a prostate presentedin this paper,we utilize theoptimal value of parameters,,and in the distance regularized level set evolutionmodel and the deformableshape prior of prostatetosegment a part of images from normal prostate, benign hyperplasiaprostate and cancer prostate.The experiment results show that the MRI segmentation method of prostate presented in this paper is effective for differentsituation of different patients
