作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2006, Vol. 32 ›› Issue (21): 209-211. doi: 10.3969/j.issn.1000-3428.2006.21.073

• 多媒体技术与应用 • 上一篇    下一篇

图像数据挖掘在SARS辅助诊断中的应用

万寿红,李 曦,龚育昌,谢铉洋   

  1. (中国科学技术大学计算机科学技术系,安徽省计算与通讯软件重点实验室,合肥 230027)
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2006-11-05 发布日期:2006-11-05

Application of Image Data Mining to
Computer Aided Diagnosis SARS

WAN Shouhong, LI Xi, GONG Yuchang, XIE Xuanyang   

  1. (Department of Computer Sci. & Tech., University of Sci. & Tech.,
    Anhui Province Key Laboratory of Software in Computing and Communication, Hefei 230027)
  • Received:1900-01-01 Revised:1900-01-01 Online:2006-11-05 Published:2006-11-05

摘要: 严重急性呼吸道综合症(SARS),又称“非典型肺炎”,是目前人类面临的一种严重危害生命和健康的新发传染病。利用PACS系统中的胸部数字X光(DX)正位图像,采用图像数据挖掘技术,设计并实现了SARS计算机辅助诊断系统。经过数据清理定位DX肺部图像的感兴趣区域,分割出双肺区域,提取特征参数,构造决策树,实现对SARS患者和一般肺炎胸部DX正位图像的分类。实验结果表明,检测SARS图像正确率达到70%以上。

关键词: 图像数据挖掘, 计算机辅助诊断, SARS, 图像分割, 决策树

Abstract: Severe acute respiratory syndrome (SARS), called “typical Pneumonia” in China, is a newly occurred fast transmittable infectious disease which badly endangers human’s life and health. This paper designs and realizes a computer aided diagnosis SARS based on image data mining techniques for digital X-Ray images in picture archiving and communication system (PACS). First, lung region of interest is located after data cleaning. Then lung region segmentation and feature parameters extraction are performed. The decision tree is constructed for discrimination of SARS and “typical Pneumonia”. The experiment result shows that more than 70% SARS cases can be detected.


Key words: Image data mining, Computer aided diagnosis(CAD), Severe acute respiratory syndrome(SARS), Image segmentation, Decision tree

中图分类号: