中文摘要：

目的研究MS患者常规磁共振图像（cMRI）上表现正常的胼胝体部位的纹理特征及其方向特性,为显示MS患者脑组织的隐匿性损伤提供新的方法。方法选取37例复发-缓解型MS患者和年龄、性别匹配的37例健康对照者图像,在cMRI上表现正常胼胝体的矢状位T1加权图像提取膝部、压部、体部感兴趣区,分别组成三个部位的MS组和对照组样本集,映射成灰度共生矩阵,然后提取能量、逆差矩、熵等纹理参量,比较各部位两组间的纹理特征差异。结果胼胝体膝部和压部在0°、45°、90°和135°四个方向,MS组能量和逆差矩值明显高于对照组,熵值明显低于对照组（P〈0.05）;体部在0°、45°和135°方向逆差矩值显著高于对照组（P〈0.05）。结论MS患者MRI的T1加权图像上表现正常胼胝体的纹理特征与健康对照组具有显著性差异,特别是膝部和压部更为显著。纹理特征分析为显示MS患者cMRI上表现正常脑组织的隐匿性病变提供了新方法。

英文摘要：

Objective The aim of this study is to investigate the texture characters and its directional feature of normal appearance corpus callosum in conventional magnetic resonance images （cMRI） from patients with multiple sclerosis （MS）, and to explore a new method for revealing abnormality underlying the brain tissue of MS. Methods T1 weighted MR images of sagittal from thirty-seven patients with remitting-relapsing multiple sclerosis and another thirty-seven volunteers matched with age and gender were selected. Regions of interest （ROD from three callosal parts, genu, body and splenium, were extracted from normal appearance corpus callosum individually and correspondingly divided into three sections, each with MS group and control group. All of the images were mapped into gray level co-occurrence matrix and then texture characteristics of energy, inverse difference moment and entropy were calculated in order to test the significance between two groups. Resuits The values of energy and inverse difference moment in genu and splenium of corpus callosum in MS groups were significantly higher than that in control groups with all the four directions, 0^o, 45^o, 90^o and 135^o and the value of entropy were reversed（P〈0.05） spontaneously with significance. The value of inverse difference moment of MS group was significantly higher than that of the controls in the body part with the direction of 0^o, 45^o and 135^o（P〈0.05）. Conclusion Significant differences of texture characters on T1 weighted MR image in sagittal are existed between MS groups and the controls, especiallly in the parts of genu and splenium. Texture analysis may provide some new methods in revealing the hidden abnormality of normal appearing brain tissues with cMRI of MS.