Liver segmentation for CT images using an improved GGVF-snake
Tianyi Gui, Lin-Lin Huang, Akinobu Shimizu · 2007
Accurate liver segmentation from abdominal computed tomography (CT) images is one of the most important steps for computer aided diagnosis (CAD) for liver CT. In this paper, we present a hybrid method for semiautomatic delineation of the liver contours on CT images. Firstly, the CT images are enhanced and denoised by a method based on histogram equalization and anisotropic diffusion filtering; Then, a manually delineated boundary using hermite-spline interpolation is chosen as the rough segmentation result; Finally, an improved generalized gradient vector flow snake model (GGVF-Snake) based on canny algorithm is adopted for refinement of the rough segmentation. Experiment results show that the proposed method can precisely extract the liver region.