样条几何形变模型及其在CT图像分割上的应用

根据医学图像分割的特点，建立了一种样条几何形变模型，它表示为一个有序的用边线连接的顶点集合，其有序的顶点集合组成顶点模型，有序的边线集合组成边线模型，两者通过形变模型的二阶连续性紧密地结合在一起，样条几何形变模型能够获得平滑而连续的对象描述，适合于形状复杂的解剖结构，将其用于肝脏CT图像分割，结果表明，模型具有很好的收敛性和描述效果。

A Spline Geometric Deformable Model and Its Application in Liver Segmentation from CT Images

Following the characteristics of medical image segmentation, this paper proposed a new spline geometric deformable model (SGDM), which can be presented as a set of ordered vertices connected by edges. The ordered vertices form its vertex model, and its ordered edges form the edge model, both are integrated together by the continuities of the deformable model. The new deformable model can provide a smooth and continuous representation of anatomical object with complicated shape. It has been applied to the segmentation of CT images for the liver, and the results show good convergence and representation of object's shape and complexity.