计算机应用 ›› 2020, Vol. 40 ›› Issue (8): 2455-2459.DOI: 10.11772/j.issn.1001-9081.2019122105

• 应用前沿、交叉与综合 • 上一篇    下一篇

基于MRI图像的阿尔茨海默症患者脑网络特征识别算法

朱琳1, 于海涛1, 雷新宇1, 刘静2, 王若凡3   

  1. 1. 天津大学 电气自动化与信息工程学院, 天津 300072;
    2. 唐山市工人医院 神经内科, 河北 唐山 063000;
    3. 天津职业技术师范大学 信息技术工程学院, 天津 300222
  • 收稿日期:2019-12-13 修回日期:2020-03-05 出版日期:2020-08-10 发布日期:2020-08-21
  • 通讯作者: 王若凡(1986-),女,河北沧州人,讲师,博士,主要研究方向:神经信息处理,wangrf@tju.edu.cn
  • 作者简介:朱琳(1996-),女,天津人,硕士研究生,主要研究方向:脑认知、神经信息处理;于海涛(1985-),男,河北唐山人,副教授,博士生导师,博士,主要研究方向:脑认知、神经信息处理;雷新宇(1995-),男,河北唐山人,硕士研究生,主要研究方向:脑认知、神经信息处理;刘静(1973-),女,河北唐山人,博士,主要研究方向:神经病学。
  • 基金资助:
    国家自然科学基金青年科学基金资助项目(61601331);天津市自然科学基金资助项目(19JCYBJC18800,18JCQNJC04700);河北省重点研发项目(18277773D)。

Brain network feature identification algorithm for Alzheimer's patients based on MRI image

ZHU Lin1, YU Haitao1, LEI Xinyu1, LIU Jing2, WANG Ruofan3   

  1. 1. School of Electrical and Information Engineering, Tianjin University, Tianjin 300072, China;
    2. Department of Neurology, Tangshan Gongren Hospital, Tangshan Hebei 063000, China;
    3. School of Information Technology and Engineering, Tianjin University of Technology and Education, Tianjin 300222, China
  • Received:2019-12-13 Revised:2020-03-05 Online:2020-08-10 Published:2020-08-21
  • Supported by:
    This work is partially supported by the Youth Program of National Natural Science Foundation of China (61601331), the Tianjin Natural Science Foundation (19JCYBJC18800, 18JCQNJC04700), the Hebei Key Research and Development Project (18277773D).

摘要: 针对通过脑成像对阿尔茨海默症(AD)进行人工识别存在主观性、易误诊的问题,提出了一种基于核磁共振成像(MRI)图像构建脑网络对AD进行自动识别的方法。首先,把MRI图像叠加并进行结构块划分,并通过计算任意两个结构块之间的结构相似性(SSIM)来构造网络;然后,利用复杂网络理论提取结构参数,并将其作为机器学习算法的输入实现AD的自动识别。分析发现双参数特别是节点介数和边介数作为输入时分类效果最优,进一步研究发现MRI图像划分为27个结构块时分类效果最优,对于加权网络和无权网络的准确率分别最高可达91.04%和94.51%。实验结果表明,基于MRI结构块划分构建的结构相似性复杂网络能够对AD进行准确率更高的识别。

关键词: 脑网络, 磁共振成像, 机器学习, 阿尔茨海默症, 结构相似性

Abstract: In view of the problem of subjectivity and easy misdiagnosis in the artificial identification of Alzheimer's Disease (AD) through brain imaging, a method of automatic identification of AD by constructing brain network based on Magnetic Resonance Imaging (MRI) image was proposed. Firstly, MRI images were superimposed and were divided into structural blocks, and the Structural SIMilarity (SSIM) between any two structural blocks was calculated to construct the network. Then, the complex network theory was used to extract structural parameters, which were used as the input of machine learning algorithm to realize the AD automatic identification. The analysis found that the classification effect was optimal with two parameters, especially the node betweenness and edge betweenness were taken as the input. Further study found that the classification effect was optimal when MRI image was divided into 27 structural blocks, and the accuracy of weighted network and unweighted network was up to 91.04% and 94.51% respectively. The experimental results show that the complex network of structural similarity based on MRI block division can identify AD with higher accuracy.

Key words: brain network, Magnetic Resonance Imaging (MRI), machine learning, Alzheimer's Disease (AD), structural similarity

中图分类号: