Background: Intensity in homogeneity occurs in many medical images, especially in\nvessel images. Overcoming the difficulty due to image in homogeneity is crucial for the\nsegmentation of vessel image.\nMethods: This paper proposes a localized hybrid level-set method for the\nsegmentation of 3D vessel image. The proposed method integrates both local region\ninformation and boundary information for vessel segmentation, which is essential for\nthe accurate extraction of tiny vessel structures. The local intensity information is firstly\nembedded into a region-based contour model, and then incorporated into the\nlevel-set formulation of the geodesic active contour model. Compared with the preset\nglobal threshold based method, the use of automatically calculated local thresholds\nenables the extraction of the local image information, which is essential for the\nsegmentation of vessel images.\nResults: Experiments carried out on the segmentation of 3D vessel images\ndemonstrate the strengths of using locally specified dynamic thresholds in our level-set\nmethod. Furthermore, both qualitative comparison and quantitative validations have\nbeen performed to evaluate the effectiveness of our proposed model.\nConclusions: Experimental results and validations demonstrate that our proposed\nmodel can achieve more promising segmentation results than the original hybrid\nmethod does.
Loading....