一种优化树状模型的肝脏自动分割方法

摘要

针对CT 图像的复杂性与肝脏形态的多样性,提出一种优化的肝脏自动分割方法.通过基于部件共享池的混合树状模型(TSPM)捕获肝脏边界的拓扑形态变化,利用凹凸点算法根据肝脏形变自动筛选TSPM 中的关键点,避免冗余点对肝脏边界的错误定位,并将不同颜色空间应用于肝脏图像分割中提高分割精度.实验结果表明,与现有分割方法相比,优化方法可获得更准确的肝脏分割结果.%Aiming at the complexity of CT images and the diversity of liver morphology,an automatic optimized liver segmentation method is proposed. Through a hybrid tree model based on the shared pool of components,called TSPM,the topological changes of the liver boundary can be captured. The proposed Convex Concave Point (CCP) algorithm can automatically screen out key points to avoid the misalignment of redundant points on the liver boundary. It applys different color spaces to liver image segmentation to improve segmentation accuracy. Experimental results show that compared with the existing segmentation method,the optimized method can obtain more accurate liver segmentation results.