计算机应用 ›› 2013, Vol. 33 ›› Issue (06): 1737-1741.DOI: 10.3724/SP.J.1087.2013.01737

• 多媒体技术 • 上一篇    下一篇

分割多发性硬化症白质病灶的新方法

相艳,贺建峰,马磊,易三莉,徐家萍   

  1. 昆明理工大学 信息工程与自动化学院,昆明650051
  • 收稿日期:2012-12-28 修回日期:2013-02-13 出版日期:2013-06-01 发布日期:2013-06-05
  • 通讯作者: 相艳
  • 作者简介:相艳(1979-),女,云南大理人,讲师,硕士,主要研究方向:医学图像处理;贺建峰(1965-),男,云南开远人,教授,博士,主要研究方向:医学图像处理;马磊(1978-),男(回族),云南昆明人,讲师,硕士,主要研究方向:信号处理、软件工程;易三莉(1977-),女,湖南岳阳人,讲师,博士,主要研究方向:医学图像处理;徐家萍(1974-),女,云南昭通人,讲师,硕士,主要研究方向:生物医学信号处理。
  • 基金资助:

    国家自然科学基金资助项目(60872092);教育部回国人员科研启动基金资助项目(2010-1561)

New method for multiple sclerosis white matter lesions segmentation

XIANG Yan,HE Jianfeng,MA Lei,YI Sanli,XU Jiaping   

  1. Faculty of Information Engineering and Automation, Kunming University of Science and Technology, Kunming Yunnan 650051, China
  • Received:2012-12-28 Revised:2013-02-13 Online:2013-06-05 Published:2013-06-01
  • Contact: XIANG Yan

摘要: 多发性硬化症(MS)是一种慢性的中枢神经系统疾病,其病灶可由常规脑部核磁共振成像(cMRI)进行检测。为提高图像处理的效率,提出了一种自动分割cMRI图像中的MS白质病灶(WML)的新方法。首先将模糊核聚类(KFCM)用于预处理后的T1加权像,得到白质图像;然后利用一个种子点的区域生长处理白质图像,提取出一个二值模板。该模板与对应的T2加权像进行乘积,得到一幅仅包含白质、病灶及背景的图像;最后再次利用KFCM分割图像,得到病灶的核心部分。实验结果表明,所提出的方法能快速、有效地分割出低噪声仿真图像中的WML,且Dice相似性系数平均值在80%以上。

关键词: 多发性硬化症, 模糊核C-均值聚类, 常规磁共振成像, 分割, 白质

Abstract: Multiple Sclerosis (MS) is a chronic disease that affects the central nervous system and MS lesions are visible in conventional Magnetic Resonance Imaging (cMRI). A new method for the automatic segmentation of MS White Matter Lesions (WML) on cMRI was presented, which enabled the efficient processing of images. Firstly the Kernel Fuzzy C-Means (KFCM) clustering was applied to the preprocessed T1-weight (T1-w) image for extracting the white matter image. Then region growing algorithm was applied to the white matter image to make a binary mask. This binary mask was then superimposed on the corresponding T2-weight (T2-w) image to yield a masked image only containing white matter, lesions and background. The KFCM was reapplied to the masked image to obtain WML. The testing results show that the proposed method is able to segment WML on simulated images of low noise quickly and effectively. The average Dice similarity coefficient of segmentation result is above 80%.

Key words: multiple sclerosis, Kernel Fuzzy C-Means (KFCM) clustering, conventional Magnetic Resonance Imaging (cMRI), segmentation, white matter

中图分类号: