中文摘要：

目的 分析阿尔茨海默（Alzheimer＇s disease,AD）、轻度认知障碍（mild cognitive impairment,MCI）和正常对照者（normal controls,NC）MR图像扣带纹理特征,并按性别进行分组分析,探索纹理特征在疾病早期诊断上的应用.方法 利用灰度共生矩阵和游程长矩阵对52例MR图像（AD14例、MCI20例、NC18例）进行纹理分析,测试组间参数是否显著不同,并用支持向量机方法对在组间具有显著性差异的纹理特征进行分类识别.结果 与NC组相比,AD和MCI组的能量、游程长不均匀度等纹理特征在前扣带和后扣带均存在显著性差异.按性别分组的实验结果显示,除AD组和NC组间存在显著不同外,男性后扣带在MCI与NC及MCI与AD间均存在显著性差异（P〈0.05）,女性前扣带在MCI和NC组间存在显著性差异（P〈0.05）.分类识别结果显示,男性后扣带分类正确率最高,为90%.结论 MR图像纹理特征可以反映扣带病理变化,不同性别MR图像表现不同,有可能为AD的早期诊断提供帮助.

英文摘要：

Objective This article studied the texture features of cingulum of patients with Alzheimer disease（AD） and mild cognitive impairment （MCI）based on magnetic resonance images and analyzed them according to different gender. The purpose was to explore the application of texture features in early diagnosis of AD. Methods Texture analysis was applied to MR images of 14 AD,20 MCI and 18 normal controls（NC）. Texture features extracted from gray level co-occurrence matrix and run-length matrix were analyzed between each two groups. The texture features that existed significant differences among groups were used as input features in a classification procedure based on support vector machines （SVM）. Results Compared with NC group,texture features including energy and run length non-uniformity of AD and MCI group were significantly different on anterior and posterior cingulum. The results grouped by gender showed that except the differences between AD and NC group,male posterior cingulum showed significant differences between MCI and NC group and MCI and AD group （P 〈 0.05）, female anterior cingulum showed significant differences between MCI and NC group（P 〈 0.05）. The classification accuracy for AD and NC group of male posterior cingulum was 90%. Conclusion The results indicated that the pathological changes in cingulum could be reflected by texture features and the pathological changes may be different in the two genders. This may be helpful to early diagnosis of AD.