北京生物医学工程

基于纹理分析的阿尔茨海默症及轻度认知功能障碍的分类研究

Classification of Alzheimer disease and mild cognitive impairment from normal controls based on texture analysis

作者：刘卫芳夏翃王旭周震

单位：                      首都医科大学生物医学工程学院（北京100069）        

关键词：阿尔茨海默症；轻度认知功能障碍；三维纹理分析；胼胝体；分类识别

分类号：

出版年·卷·期（页码）：2014·33·6（609-613）

摘要：

目的利用脑MR图像中胼胝体的三维纹理特征对阿尔茨海默症患者（Alzheimer disease, AD）及轻度认知功能障碍（mild cognitive impairment, MCI）患者进行分类识别，以探索AD 早期诊断新途径。方法选取AD患者、MCI患者及健康对照者各18例，采用灰度共生矩阵和游程长矩阵提取每位受试者胼胝体部位的三维纹理特征。通过筛选得到的纹理特征参量，利用BP神经网络建立识别模型，对AD患者、MCI患者和健康对照者进行分类识别，并对采用主成分分析、线性判别分析和非线性判别分析3种方法得到的识别结果进行比较。结果使用神经网络模型的非线性判别分析的分类识别正确率最高。结论利用三维纹理特征的神经网络模型可分类识别早期AD患者及MCI患者。

Objective This study investigated three-dimensional (3D) texture as a possible diagnostic marker of Alzheimer disease. Methods T1-weighted MR images were obtained from 18 AD patients, 18 MCI patients and 18 age and gender-matched normal controls. 3D texture features of the corpus callosum were extracted from the gray level co-occurrence matrix and run length matrix. The texture features that existed significant differences among the three groups were used as features in a classification procedure. Results The classification accuracy of nonlinear discriminant analysis was the highest with neural network model. Conclusions Three-dimensional texture can recognize the pathological changes of corpus callosum in patients with AD and MCI, and might be a useful aid in AD diagnosis.

参考文献：

服务与反馈：

【文章下载】【加入收藏】

提示：您还未登录，请登录！点此登录