目的利用双源CT图像对心脏二尖瓣瓣膜进行了增强与分割,为日后进行二尖瓣三维重建和运动分析提供基础。方法提出了自适应种子填充法和Contourlet小波变换分别用于不同情况下的分割:首先用自适应种子填充法对双源CT胸腔图像进行心脏在造影剂下的区域分割,再用Contourlet小波变换对区域分割后的图像进行二尖瓣的增强和分割。结果在MATLAB7.0的环境下对双源CT图像成功进行了处理和分割,获取了清晰的心脏二尖瓣图像。结论本文提出的分割方法对不同情况下的分割非常有效,实验的结论对医生从事二尖瓣瓣膜运动研究和二尖瓣瓣膜外科手术规划提供有益的参考。
Objective Children' s heart valves by Dual-Source Compater Tomography(DS CT) images were enhanced and segmented in order to reconstruct 3D construction of heart valves and make motion analysis based on this model. Method Adaptive seed-filling algorithm is proposed in this paper to segment interested region of heart based on DS CT images. Contourlet transform is used to enhance and segment heart valve based on the part-segmented image. Result With these methods, DS CT images were success fully processed and segmented in Matlab 7.0 and clear images of mitral valves were got. Conclusions The adaptive seed-filling algorithm proposed in this paper and Contourlet transform are very useful for segmentation in different situation. The experimental result in this paper provides useful reference for doctors in their research on the motion of mitral valves and planning on surgical operation of mitral valves.