基于分形维理论的多发性硬化患者脑白质的纹理特征分析  

Texture analysis on normal appearing white matter in patients with multiple sclerosis based on fractal theory

作　　者：王磊[1] 童隆正[1] 于春水[2] 

机构地区：[1]首都医科大学生物医学工程学院,北京100069 [2]首都医科大学宣武医院,北京100054

出　　处：《北京生物医学工程》2008年第2期134-137,共4页Beijing Biomedical Engineering

基　　金：国家自然基金项目(30670575);北京市自然科学基金项目(3073015)资助

摘　　要：多发性硬化是一种以脱髓鞘和轴索损伤为病理特征的慢性炎性疾病,可造成严重的神经功能障碍,近年来在中国的发病率有所上升。MRI是诊断多发性硬化的重要工具,在MRI上可见有形病灶,但无法检测出肉眼不可见的隐匿性损伤。本文采用基于分形维理论的纹理分析方法对患者和健康对照组的MR图像脑白质区域提取分形维纹理特征参数,然后进行统计学分析,发现患者非病灶区的脑白质和健康对照组脑白质之间存在显著性差异。此特征参数有可能检测出MS患者MR图像上表现正常脑白质内的隐匿性损伤,为MS的研究提供了新的方法。Multiple sclerosis （MS） is a chronic inflammatory disease of the central nervous system,which leads to severe progressive neurologic dysfunction, such as blindness, paralysis and death. Recently the incidence rate of MS is increasing in China. Magnetic resonance imaging （MRI） is a powerful tool for diagnosing MS. Although definite focal plaques are observed on MRI, dormant damage can not be detected. In this paper, regions of interesting were chosen in MRI T2WI as research object; we used fractal method to analyze the texture features of MRI for patients with MS. We discussed the texture features of brain white matter, extracted texture parameters and found there was significant difference between normal brain white matter and normal appearing white matter, which was proved by the statistical analysis. This study will be beneficial to detect dormant damage in patients with MS and provide a new method for researching MS.

关 键 词：分形维 纹理分析 MRI 多发性硬化 

分 类 号：R318.04[医药卫生—生物医学工程]