基于纹理的乳房超声图像中肿瘤区域自动检测

计算机工程与应用 ›› 2007, Vol. 43 ›› Issue (18): 221-224.

基于纹理的乳房超声图像中肿瘤区域自动检测

雷琴^1,许存禄²,王小平¹

1.兰州交通大学数理与软件学院,兰州 730070
2.复旦大学信息学院计算机科学与工程系,上海 200433

收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2007-06-21 发布日期:2007-06-21
通讯作者: 雷琴

Automatic detection of breast tumor regions in medical ultrasound images with texture features

LEI Qin¹,XU Cun-lu²,WANG Xiao-ping¹

1.School of Mathematics,Physics and Software Engineering,Lanzhou Jiaotong University,Lanzhou 730070,China
2.Dept. Computer Science and Engineering,College of Information,Fudan University,Shanghai 200433,China

Received:1900-01-01 Revised:1900-01-01 Online:2007-06-21 Published:2007-06-21
Contact: LEI Qin

摘要/Abstract

摘要： 纹理属性是医学图像分析的一个非常重要的方法,探索利用一种称为统计地形特征的新的纹理描述方法自动监测超声医学图像乳房肿瘤区域。统计地形特征首先用一个可变水平平面去切割图像函数图可以得到一些实体,然后利用从这些实体的几何和拓扑属性的某些统计信息提取的6条特征曲线来描述纹理。文中实验部分验证了统计地形特征在自动监测超声医学图像乳房肿瘤区域中具有良好的效果。

Abstract: Texture is a very important research work in the field of medical image analysis,this paper explores a new texture description method,statistical landscape features,to automatically detect breast tumor regions in medical ultrasound images.Statistical landscape features first uses a variable horizontal plane to cut the image graph and derives some solids.Six texture feature curves based on the statistics of geometrical and topological properties of the solids are then used to characterize the texture.Experimental result indicates that statistical landscape features shows very remarkable effects in automatic detection of breast tumor regions in medical ultrasound images.

雷琴^1,许存禄²,王小平¹. 基于纹理的乳房超声图像中肿瘤区域自动检测[J]. 计算机工程与应用, 2007, 43(18): 221-224.

LEI Qin¹,XU Cun-lu²,WANG Xiao-ping¹. Automatic detection of breast tumor regions in medical ultrasound images with texture features [J]. Computer Engineering and Applications, 2007, 43(18): 221-224.