Local Hybrid Level-set Method for MRA Image Segmentation

In this paper, a local hybrid level-set method for medical image segmentation is presented. In proposed method, a locally fitted binary energy function is introduced into the hybrid level-set framework proposed by Zhang et al.. Compared with the globally specified threshold, the use of local binary fitting energy in the hybrid level-set method allows one to extract local image information more accurately, which is essential for enhancing the efficiency and effectiveness of the segmentation of inhomogeneous images. Experimental results on 3D medical images are presented to demonstrate the strengths of the proposed method.