基于层次化子块生长的肝脏血管模型提取

Liver Vascular Model Extraction Based on the Growth of Hierarchical Sub-block

肝内血管模型的提取是实施肝脏分段和手术模拟的重要基础,为此提出基于竞争策略的三维层次化子块生长肝脏血管提取算法。由用户选择待分割的目标种子点和周围其他组织的种子点,算法自动根据种子点邻域子块的灰度值分布计算生长准则,各种子点根据生长准则进行子块生长,同时算法引入竞争策略处理目标种子点和其他种子点的生长竞争,以该子块的均值与种子点均值的差异最小选择生长种子点,以抑制目标区域的过生长。该算法对肝脏CT图像进行血管分割测试,结果表明能有效地提取出三维血管。

The extraction of intrahepatic vessels is important for liver segmentation and surgical simulation. And a liver vessels segmentation algorithm is proposed based on three-dimension hierarchical sub-block growing and competitive strategy. The seed points are chosen by the user, which are divided into target seed points and the seed points of the surroundings organizations. Then the growth criteria is automatically calculated according to the gray value distribution of the sub-blocks of the seed points. Each seed point is growing with its growth criterion. And the competition strategy is used to deal with the growth competition of the target seed point and other seed points, which the least difference among the mean value of the sub-block and the mean value of the seed points is growing. This is useful to suppress the over-growth of the target region. The algorithm is tested to segment the liver vessels based on the CT images, and the results were effective in extracting the three-dimensional vessels.