| 李慧卓,相洁,秦嘉玮,梁佩鹏,李坤成.基于Adaboost的轻度认知障碍和阿尔茨海默病分类[J].中国医学影像技术,2016,32(4):623~627 |
| 基于Adaboost的轻度认知障碍和阿尔茨海默病分类 |
| Classification of mild cognitive impairment and Alzheimer disease based on Adaboost algorithm |
| 投稿时间：2015-07-20 修订日期：2015-12-26 |
| DOI：10.13929/j.1003-3289.2016.04.037 |
| 中文关键词: 集成分类 阿尔茨海默病 轻度认知障碍 磁共振成像 |
| 英文关键词:Ensembled classification Alzheimer disease Mild cognitive impairment Magnetic resonance imaging |
| 基金项目:国家自然科学基金(61373101、61473196)、北京市科技新星项目(Z12111000250000、Z131107000413120)、北京市卫生系统高层次卫生技术人才学科骨干项目及认知神经科学与学习国家重点实验室开放课题(CNLZD1302)、山西省科技攻关项目(20140321001-01). |
|
| 摘要点击次数: 2016 |
| 全文下载次数: 761 |
| 中文摘要: |
| 目的 采用Adaboost集成分类方法区分轻度认知障碍(MCI)、阿尔茨海默病(AD)患者与正常对照(NC)的功能与结构磁共振成像数据.方法 对26例MCI患者(MCI组)、26例AD患者(AD组)及30名健康老年人(NC组)的MRI图像进行分析,选择双侧海马体积及3组间存在显著差异脑区的低频振幅值(ALFF)作为分类特征,采用Adaboost集成分类器对3组被试进行两两分类,利用留一交叉验证估算分类准确率.结果 增加性别、年龄和MMSE特征后,Adaboost集成分类方法对AD与MCI、MCI与NC、AD与NC分类准确率分别达98.08%、80.36%和100%.结论 Adaboost集成分类方法可较好地区分MCI、AD与NC. |
| 英文摘要: |
| Objective To distinguish the function and structure of MRI data about mild cognitive impairment (MCI), Alzheimer disease (AD) patients and normal controls (NC) with Adaboost integrated classification method . Methods Totally 26 MCI patients (MCI group), 26 AD patients (AD group), and 30 aged NC (NC group) were included. MRI data of the 82 subjects were analyzed by choosing the volume of the bilateral hippocampus and the amplitude of low frequency of the brain areas with significant differences as the features. The ensembled classifier Adaboost was applied to evaluate the classification accuracy between the three groups using a leave-one-out cross-validation. Results After increasing gender, age and MMSE features, the classification accuracy for AD vs MCI, MCI vs NC and AD vs NC were 98.08%, 80.36%, and 100%, respectively. Conclusion Adaboost is an effective method for differenting MCI, AD patients from NC. |
| 查看全文 查看/发表评论 下载PDF阅读器 |
|
|
|
