Computer Engineering and Applications ›› 2013, Vol. 49 ›› Issue (4): 185-188.

Previous Articles     Next Articles

Mammogram density estimation based on sub-region classification

LIU Qingqing, LIU Li, WANG Jian   

  1. School of Electronic & Information Engineering, Tianjin University, Tianjin 300072, China
  • Online:2013-02-15 Published:2013-02-18

基于子区域分类的乳腺密度估计

刘庆庆,刘  立,王  建   

  1. 天津大学 电子信息工程学院,天津 300072

Abstract: Breast density is a widely adopted measure for early breast cancer diagnose. An automated breast density estimation method is proposed. Different with the previous methods, the presented method first divides a mammogram into a set of sub-regions. Then sub-regions are classified as high density and low density categories based on their intensity distribution. The breast density in the mammogram is evaluated by calculating the ratio of number of high density sub-regions to that of the whole set. Groups of histogram moments of sub-regions are extracted as inputs of the Support Vector Machine(SVM) to classify the sub images. Experimental results show that the good performance of the proposed method.

Key words: breast density, histogram moment, sub-region classification, Support Vector Machine

摘要: 乳腺密度常用于乳腺癌早期诊断。提出了一种基于子区域分析的乳腺密度估计方法。该方法先将整幅钼靶X线图像中的乳腺区域分割为互不重叠的子区域,采用直方图矩描述各子区域的灰度分布,并结合支持向量机将各子区域分为高密度和低密度两类;通过计算高密度子区域占所有子区域的比例,最终得到钼靶图像中乳腺密度。实验表明,该方法对乳腺X线图像具有很好的分类效果。

关键词: 乳腺密度, 直方图矩, 子区域分类, 支持向量机