阿尔茨海默病(AD)是一种神经退行性疾病。随着脑医学影像的发展，对AD诊断的精确度也在进一步提高，但对AD的诊断，客观上仍缺少好的生物标记。为寻找到AD的更稳定的生物标记，利用海马的影像组学特征对海马的信号强度、形状、灰度阶梯分布等特征进行刻画，通过方差分析(ANOVA)和事后检验，在统计学上寻找出正常对照(NC)、AD、轻度认知损害(MCI)之间存在差异的特征；通过与被试的简易智能状况检查(MMSE)评分进行相关性分析，找寻与MMSE评分相关性较高的特征；利用支持向量机(SVM)构建一个对AD和NC分类的模型，交叉验证得到的正确率为86%。结果表明，海马的影像组学特征是一个很好的生物标记，能对AD进行有效的早期识别。

Alzheimer's Disease(AD) is a neurodegenerative disease. With the development of brain imaging, the accuracy of AD diagnosis has been well improved. However, there still lack of good markers to early diagnosis of AD. In order to find more stable biomarkers of AD, the hippocampus intensity, shape, grayscale gradient distribution and other characteristics are investigated by an Analysis Of Variance and post-hoc analysis so as to identify the alteration in AD and Mild Cognitive Impairment(MCI) in comparison to Normal Controls(NC). The results show that the Mini-Mental State Examination(MMSE) scores are significantly associated with the radiomic features of the bilateral hippocampus. The classification analysis with the Support Vector Machine(SVM) shows that an accuracy of 86% can be obtained with leaving one out cross validation, which indicates that the hippocampal texture might be taken as one of the potential markers for AD.