A Vessel Segmentation Method for MRA Images Based on Multi-scale Analysis and Level Set Framework

This paper presents a simple and efficient vessel segmentation algorithm for three-dimensional (3D) magnetic resonance angiography (MRA) images based on the level set framework. Level set methods represent a surface as the zero level set of an embedding hypersurface and evolve the hypersurface. The hypersurface is initialized using seed points inside vessels and evolved under the guidence of an evolution function. We use multi-scale Hessian analysis and mean curvature to construct the evolution function. The proposed method is tested on 6 real MRA images taken from different anatomical regions and 1 synthetic image. The segmentation results of the real images are evaluated visually and the synthetic image is used for quantitative evaluation.     «

This paper presents a simple and efficient vessel segmentation algorithm for three-dimensional (3D) magnetic resonance angiography (MRA) images based on the level set framework. Level set methods represent a surface as the zero level set of an embedding hypersurface and evolve the hypersurface. The hypersurface is initialized using seed points inside vessels and evolved under the guidence of an evolution function. We use multi-scale Hessian analysis and mean curvature to construct the evolution...     »