Journal of Computer Applications ›› 2017, Vol. 37 ›› Issue (8): 2410-2415.DOI: 10.11772/j.issn.1001-9081.2017.08.2410

Previous Articles     Next Articles

Dimension reduction method of brain network state observation matrix based on Spectral Embedding

DAI Zhaokun, LIU Hui, WANG Wenzhe, WANG Yanan   

  1. Faculty of Information Engineering and Automation, Kunming University of Science and Technology, Kunming Yunnan 650504, China
  • Received:2017-01-13 Revised:2017-03-08 Online:2017-08-10 Published:2017-08-12
  • Supported by:
    This work is partially supported by the National Natural Science Foundation of China (61263017),the Teaching Reform Project of Kunming University of Science and Technology (10968397).


代照坤, 刘辉, 王文哲, 王亚楠   

  1. 昆明理工大学 信息工程与自动化学院, 昆明 650504
  • 通讯作者: 刘辉
  • 作者简介:代照坤(1990-),男,陕西渭南人,硕士研究生,主要研究方向:人脑网络实时动态特性分析、高维数据降维;刘辉(1984-),男,陕西渭南人,副教授,博士,CCF会员,主要研究方向:实时计算机控制、图像处理、模式识别;王文哲(1991-),男,河南平顶山人,硕士研究生,主要研究方向:图像处理;王亚楠(1990-),女,河南信阳人,硕士研究生,主要研究方向:高维数据降维、数据挖掘。
  • 基金资助:

Abstract: As the brain network state observation matrix based on functional Magnetic Resonance Imaging (fMRI) reconstruction is high-dimensional and characterless, a method of dimensionality reduction based on Spectral Embedding was presented. Firstly, the Laplacian matrix was constructed from the similarity measurement between the samples. Secondly, in order to achieve the purpose of mapping (reducing dimension) datasets from high dimension to low dimension, the first two main eigenvectors were selected to construct a two-dimensional eigenvector space through Laplacian matrix factorization. The method was applied to reduce the dimension of the matrix and visualize it in two-dimensional space, and the results were evaluated by category validity indicators. Compared with the dimensionality reduction algorithms such as Principal Component Analysis (PCA), Locally Linear Embedding (LLE), Isometric Mapping (Isomap), the mapping points in the low dimensional space got by the proposed method have obvious category significance. According to the category validity indicators, compared with Multi-Dimensional Scaling (MDS) and t-distributed Stochastic Neighbor Embedding (t-SNE) algorithms, the Di index (the average distance among within-class samples) of the proposed method was decreased by 87.1% and 65.2% respectively, and the Do index (the average distance among between-class samples) of it was increased by 351.3% and 25.5% respectively. Finally, the visualization results of dimensionality reduction show a certain regularity through a number of samples, and the effectiveness and universality of the proposed method are validated.

Key words: dimensionality reduction of high dimensional data, functional brain network, brain network, state observation matrix, Spectral Embedding algorithm, dynamic characteristics

摘要: 针对基于功能核磁共振(fMRI)重构的脑网络状态观测矩阵维数过高且无特征表现的问题,提出一种基于谱特征嵌入(Spectral Embedding)的降维方法。该方法首先计算样本间相似性度量并构造拉普拉斯矩阵;然后对拉普拉斯矩阵进行特征分解,选取前两个主要的特征向量构建2维特征向量空间以达到数据集由高维向低维映射(降维)的目的。应用该方法对脑网络状态观测矩阵进行降维并可视化在二维空间平面,通过量化类别有效性指标对可视化结果进行评价。实验结果表明,与主成分分析(PCA)、局部线性嵌入(LLE)、等距映射(Isomap)等降维算法相比,使用该方法得到的脑网络状态观测矩阵低维空间的映射点有明显的类别意义表现,且在类别有效性指标上与多维尺度分析(MDS)和t分布随机邻域嵌入(t-SNE)降维算法相比,同一类样本间平均距离Di指数分别降低了87.1%和65.2%,不同类样本间平均距离Do指数分别提高了351.3%和25.5%;在多个样本上的降维可视化结果均有一定的规律性体现,该方法的有效性和普适性得以验证。

关键词: 高维数据降维, 功能脑网络, 脑网络, 状态观测矩阵, 谱特征嵌入算法, 动态特性

CLC Number: