首页 | 本学科首页   官方微博 | 高级检索  
     检索      

基于形态学和动态规划的冠状动脉树分割方法
引用本文:许燕,胡广书,耿进朝,商丽华.基于形态学和动态规划的冠状动脉树分割方法[J].清华大学学报(自然科学版),2006,46(12):2061-2064.
作者姓名:许燕  胡广书  耿进朝  商丽华
作者单位:1. 清华大学,生物医学工程系,北京,100084
2. 清华大学,第一附属医院,北京,100016
摘    要:为辅助医生临床诊断,提出了一种冠状动脉血管树的提取方法。提取方法包含基于形态学的冠状动脉路径线提取;基于圆模板的冠状动脉段的区域划分;基于动态规划方法的冠状动脉树边缘提取。该方法可以在无人工干预的情况下自动获取完整、单连通和单像素的冠脉树边缘。实验结果表明,与已有的获取冠状动脉树边缘方法相比较,该方法有很好的准确性。

关 键 词:图像处理  冠状动脉树  形态学  动态规划  边缘  Watershed(分水岭)算法
文章编号:1000-0054(2006)12-2061-04
修稿时间:2005年12月26

Automated coronary arterial tree extraction using morphological and dynamic programming segmentation
XU Yan,HU Guangshu,GENG Jinzhao,SHANG Lihua.Automated coronary arterial tree extraction using morphological and dynamic programming segmentation[J].Journal of Tsinghua University(Science and Technology),2006,46(12):2061-2064.
Authors:XU Yan  HU Guangshu  GENG Jinzhao  SHANG Lihua
Abstract:Vessel borders need to be accurately identified to aid clinical diagnoses for cardiovascular disease.An automated morphological dynamic programming method was developed to identify the coronary arterial wall.The method includes path extraction using morphological segmentation,region partitioning using a circle template,and edge detection using dynamic programming.Experimental tests gave very good results.The results show that the method more effectively detects integrated and single-pixel edges of coronary arterial trees than previous methods with no artificial intervention automatically.
Keywords:image processing  coronary arterial tree  mathematical morphology  dynamic programming  edge  Watershed method
本文献已被 CNKI 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号