Multiphase level set with multi dynamic shape models on kidney segmentation of CT image

In this paper, a multiphase level set method with multi dynamic shape models is proposed to segment the kidneys on the abdominal computed tomography (CT) images. Comparing with the original Chan-Vese model three changes are made to improve the segmentation result. The first is using shape model to help the segmentation. The second is using dynamic shape model to deal with the variation of the kidneys. The third is using multi level set to simultaneously segment multi objects. We also develop an algorithm to automatically get the initial level set curves and initial shape models which are essential to apply the proposed method. In the experiments, the proposed method is compared with the Chan-Vese model and the single level set method with shape prior to prove that the proposed method can work better on the kidneys segmentation.