|本期目录/Table of Contents|

[1]滕升华,商胜楠△,王芳,等.一种基于复合稀疏表示的阿尔茨海默病的诊断方法[J].生物医学工程研究,2016,01:7-11.
 TENG Shenghua,SHANG Shengnan,WANG Fang,et al.A Diagnosis Method for Alzheimer’s Disease based on Hybrid Sparse Representation[J].Journal of Biomedical Engineering Research,2016,01:7-11.
点击复制

一种基于复合稀疏表示的阿尔茨海默病的诊断方法(PDF)

《生物医学工程研究》[ISSN:1006-6977/CN:61-1281/TN]

期数:
2016年01期
页码:
7-11
栏目:
出版日期:
2016-03-25

文章信息/Info

Title:
A Diagnosis Method for Alzheimer’s Disease based on Hybrid Sparse Representation
作者:
滕升华1商胜楠1△王芳1赵增顺12
1.山东科技大学电子通信与物理学院,青岛 266590;2.山东大学控制科学与工程学院,济南 250061
Author(s):
TENG Shenghua1 SHANG Shengnan1 WANG Fang1 ZHAO Zengshun12
1.College of Electronics, Communication and Physics, Shandong University of Science and Technology, Qingdao 266590, China; 2.School of Control Science and Engineering, Shandong University, Jinan 250061,China
关键词:
阿尔茨海默病灰质密度图复合稀疏表示近邻传播聚类增广拉格朗日乘子法
Keywords:
Alzheimer′s disease Gray matter density map Hybrid sparse representation Affinity propagation clustering Augmented Lagrange multiplier
分类号:
R318
DOI:
-
文献标识码:
A
摘要:
利用磁共振影像数据实现对阿尔茨海默病的准确诊断。将常规稀疏表示中的单层字典分解为两层,分别使用各类别的典型样本和类内差异作为两层字典的元素;设计一种两层字典协调工作的复合稀疏表示形式,以期利用训练样本更为精确地表示待识别样本,并构建分类器用于阿尔茨海默病的分类识别。在ADNI数据库的对比实验表明,该方法的识别性能优于支持向量机和同类的稀疏表示分类器。
Abstract:
Given magnetic resonance imaging data, the accurate diagnosis method for Alzheimer′s disease was studied in this paper. By expanding the conventionally single-layer dictionary in sparse representation into two layers, the exemplars in all classes and the intra-class variations were used to construct a two-layer dictionary. Then a hybrid sparse representation was designed collaboratively incorporating the two layers of the dictionary, where the testing samples were expected to be represented more accurately with the training data. Finally a classifier based on this hybrid sparse representation was developed to diagnose Alzheimer′s disease. Experiments on Alzheimer′s Disease Neuroimaging Initiative (ADNI) dataset show promising results compared with support vector machine and traditional sparse representation based classifier.

参考文献/References

-

备注/Memo

备注/Memo:
(收稿日期:2015-10-13)国家自然科学基金资助项目(61174190,61471225);山东省自然科学基金资助项目(ZR2014FM002);中国博士后科学基金特别资助项目(2015T80717)通讯作者 Email:sdust_au@foxmail.com
更新日期/Last Update: 2016-06-13