计算机工程与应用 ›› 2013, Vol. 49 ›› Issue (18): 125-129.

• 图形图像处理 • 上一篇    下一篇

色多普勒超声肾动脉血流信号曲线分类研究

张本奎1,2,唐  娉1,李宏益1,张晓东3,李建初3   

  1. 1.中国科学院 遥感应用研究所,北京 100101
    2.中国科学院大学,北京 100039
    3.中国医学科学院 北京协和医院超声诊断科,北京 100730
  • 出版日期:2013-09-15 发布日期:2013-09-13

Research on classification of blood signal curves of renal artery from CDS

ZHANG Benkui1,2, TANG Ping1, LI Hongyi1, ZHANG Xiaodong3, LI Jianchu3   

  1. 1.Institute of Remote Sensing Applications, Chinese Academy of Sciences, Beijing 100101, China
    2.University of Chinese Academy of Science, Beijing 100039, China
    3.Department of Ultrasound, Peking Union Medical College Hospital of the Chinese Academy of Medical Sciences, Beijing 100730, China
  • Online:2013-09-15 Published:2013-09-13

摘要: 彩色多普勒超声是肾动脉狭窄的首选筛查工具,目前临床上主要依靠人工判别来诊断肾动脉狭窄,对操作者具有很强的依赖性。在肾动脉多普勒超声图像的基础上,通过提取肾动脉血流信号曲线、提取曲线特征,继而基于SVM构建分类器,对肾动脉血流信号曲线进行分类,取得了较高的分类精度,并与最大似然分类器进行了分类实验比较,在肾动脉狭窄的计算机辅助诊断方向进行了有意义的探索。

关键词: 多普勒超声, 肾动脉血流信号曲线, 支持向量机(SVM), 分类

Abstract: CDS(Color Doppler Sonography) is the first choice to screen RAS(Renal Artery Stenosis). Currently, to diagnose RAS clinically mainly relies on manual evaluation, which has a great dependency on operators. Blood signal curves of the renal artery and features of classification are extracted from CDS images, and then classifier is created based on SVM to classify blood signal curves of the renal artery with a high accuracy of classification. Besides, the result of SVM classifier is compared with that of maximum likelihood classifier. It has positive effect for diagnosing RAS with computer-aided diagnosis.

Key words: Color Doppler Sonography(CDS), blood signal curve of renal artery, Support Vector Machine(SVM), classification