肝脏CT图像三维分割研究

3D Segmentation Research of CT Liver Image Sequences

针对肝脏CT图像的特征,提出了一种将种子区域生长算法和改进Snake模型相结合的策略,实现了肝脏的三维分割提取。该方法先从CT图像序列中筛选出肝脏有明显成像边缘的一张切片,在其肝脏区域内选择若干个种子点,利用种子区域生长算法得到初始边缘,再利用改进的Snake模型对初始边缘进行优化,然后,将此切片的边缘轮廓作为与其相邻切片上的初始边缘,重复该过程,直到分割完所有切片。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合适的数据集。

By considering the characters of liver CT image sequences, a new segmentation way is proposed based on an advanced Snake model associated with seeded region growing method. With this method, a CT image clear edges is selected from CT liver image sequences, then several seeds are selected in the liver region, the seeded region growing method is used to get an initial edge. The edge is optimized by advanced Snake model and served as the initial edge of the next CT image. This procedure is repeated until all images are processed. Experimental results show that the algorithm can obtain segmentation result of soft tissue image efficiently and accurately.