中文摘要：

目的研究轻度认知功能障碍（Mild Cognitive Impairment，MCI）患者胼胝体MR图像的三维纹理特征，以期尽早发现和筛选出阿尔茨海默病（Alzheimer’s Disease，AD）的高危人群，为AD的治疗提供最佳的时间。方法选取18例MCI患者（MCI组）及18例健康对照组（NC组）的MR图像进行三维纹理分析，采用灰度共生矩阵和游程长矩阵提取每位受试者胼胝体的10个三维纹理特征，比较两组样本的各纹理特征问是否存在显著性差异。结果提取的10个纹理参数中除灰度不均匀度因子，两组间没有显著性差异（P〉0．05），其余9个参数两组间均存在显著性差异（P〈0．05）。结论三维纹理特征能反映出MCI患者胼胝体部位的病理病变，有助于AD的早期诊断和治疗。

英文摘要：

Objective 3D texture features of corpus caUosum in patients with mild cognitive impairment （MCI） based on MR images were researched to early detect and screen patients with Alzheimer＇s Disease （AD）, and offer the optimal treatment time. Methods 3D texture analysis was performed on 18 MCI patients and 18 normal controls （NC）. The 3D texture features extracted from gray level co- occurrence matrix and run length matrix were analyzed. The significant differences between the two groups were tested. Results The results showed that the texture features had significant differences （P〈0.05） between MCI and NC except grey level nonuniformity that the difference was not significant （P〉0,05） .Conclusion The microcosmic pathological changes in corpus callosum in MCI patients could be reflected by 3D texture features and may be helpful to AD early diagnosis and treatment.