基于水平集方法的软组织图像序列分割

医学图像分割是医学图像处理中的关键问题之一.图像序列的分割操作是医学图像三维重建的必要准备,而软组织图像分割则是医学图像分割中的一大难点.基于曲线演化理论的,借助偏微分方程等数学工具的水平集方法已经被广泛应用于医学图像分割领域.介绍了水平集方法的数学模型,并设计了一种基于窄带水平集方法的,专门针对软组织图像分割的算法.用边界追踪等方法提取第一层图片中的软组织相关轮廓;将它们作为初始水平集曲线,再利用窄带水平集方法进行演化;经过两个阶段的迭代处理,最终自动分割出整个软组织图像序列.实验表明该算法具有较高效率、分割结果精确,所产生的分割结果可以作为三维重建的合适的数据集.

Soft tissue image sequences segmentation based on level set method

The medical image segmentation,one of the key problems in medical image processing.The segmentation of image sequences is the necessary preparation for 3D reconstruction,and the soft tissue image segmentation is a difficultial problem in the image segmention domain.The level set method based on curves evolving theory and partial differential equation theory is widely applied in the segmentation of medical image.The model of level set method is introduced,and then an algorithm is proposed to address the problem of soft tissue image based on the narrow band level set method.First,get the first original image's edges by using edge trace method.Then make those edge curves as the initial curves for the narrow level set curves evolving.Finally,after two phase of iterative processing,the soft tissue image sequence is segmented automatically.Experimental results show that the algorithm can obtain segmentation result of soft tissue image efficiently and accurately,which can be made the proper data set for 3D reconstruction.