计算机工程与应用 ›› 2007, Vol. 43 ›› Issue (33): 221-223.

• 工程与应用 • 上一篇    下一篇

医学图像中微细管道结构的提取

程 明,黄晓阳,王博亮   

  1. 厦门大学 计算机科学系,福建 厦门 361005
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2007-11-21 发布日期:2007-11-21
  • 通讯作者: 程 明

Extraction of tenuous vasculature in medical images

CHENG Ming,HUANG Xiao-yang,WANG Bo-liang   

  1. Department of Computer Science,Xiamen University,Xiamen,Fujian 361005,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2007-11-21 Published:2007-11-21
  • Contact: CHENG Ming

摘要: 以肝动脉提取为例,提出一种医学图像中微细管道结构的提取算法。算法主要步骤包括:最大亮度投影、二维管道提取、三维种子点生成和三维管道提取。算法对常规的区域生长法进行改进,提取效果与常规的区域生长法相比有较大改善。算法需要的人工干预很少,只需要指定四个参数。

关键词: 医学图像处理, 最大亮度投影, 区域生长

Abstract: This paper proposes a method for the extraction of tenuous vasculature in medical images.The method includes the following steps:Maximum Intensity Projection(MIP),two-dimensional vasculature extraction,three-dimensional seed point generation,and three-dimensional vasculature extraction.The routine region growing algorithm is revised and the extraction result is improved.Little artificial interference is needed during the processing,except that four parameters should be designated.

Key words: medical image processing, Maximum Intensity Projection(MIP), region growing