复杂网络主成分分析的分类方法在音乐家白质可塑性研究中的应用

doi:10.3969/j.issn.0258-8021.2015.02.008

摘要
图/表
参考文献
相关文章 (5)

全文: PDF (2333 KB) HTML (1 KB)
输出: BibTeX | EndNote (RIS)

摘要人脑在多种时间和空间尺度上都是复杂网络，而复杂网络中往往包含着大量的连接信息。主成分分析（PCA）方法主要被用于从大量信息中提取重要特征，因而可以被用于探寻复杂网络中的一些重要信息。众所周知，音乐家是研究训练导致的脑可塑性问题的一个理想模型，探求音乐家脑网络的可塑性变化是非常有意义的。首先通过基于弥散加权成像（DWI）数据的纤维束追踪，构建了16位音乐家与16位非音乐家的脑白质结构网络；然后对两组人的整体脑网络进行了PCA分析，进而对得到的每个主成分做支持向量机（SVM）分类处理，得到分类效果最好的主成分；最终找出对此主成分贡献前1% 的连接即为音乐家相对于非音乐家在大脑白质结构网络上发生改变的主要连接。本方法为组间复杂网络对比分析提供了一种基于PCA分类的新思路。基于上述思路，对于音乐家与非音乐家的脑白质结构网络对比分析，表明音乐家在运动、听觉、情绪和记忆等功能脑区表现出更高的脑区间信息传递效率；进而扩展了在网络层面对长期音乐训练改变音乐家白质可塑性问题的理解。
〖HJ0〗〖HJ〗
〖HTH〗关键词：〖HTSS〗〖BP(〗┣┣（中）关键词┫┫〖BP)〗〖HT〗

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	李建福罗程^* 董立尧德中^#

关键词 ：主成分分析（PCA）, 支持向量机（SVM）, 弥散加权成像（DWI）, 脑网络

Abstract：The human brain is a complex network with multiple scales of time and space which includes large amount of connection information. Principal component analysis (PCA) can extract important features from vast quantities of information; therefore, it was used to explore important information from complex network in this study. As is widely known, musicians represent an ideal model to investigate experiencedriven plasticity changes in the human brain. It is a far more significant research that explores plasticity changes in brain networks of musicians. In this study, white matter brain networks of 16 musicians and 16 nonmusicians were firstly constructed by fiber tracking based on diffusionweighted imaging (DWI); secondly, PCA process was used to extract the feature networks of two the groups, support vector machine (SVM) classification method was then applied to each component, the component with best classification performance was obtained; finally, the first 1% connections with highest contribution to the component were considered to be the main connections which may represent the changes in the musicians’ white matter anatomical networks compared to nonmusicians. This method provides a new approach which utilizes the PCA based classification for complex network comparison issues. And, comparison analysis of the white matter anatomical brain network between musicians and nonmusicians indicated that musicians showed enhanced information transfer efficient between motor, auditory, emotional, and memory related brain regions. These findings may extend the network level understanding of white matter plasticity in musicians who have had longterm musical training.

Key words： principal component analysis (PCA) support vector machine (SVM) diffusionweighted imaging (DWI) brain network

基金资助:国家自然科学基金（91232725，81201159）；中央高校基本科研业务费项目（ZYGX 2011J097）

引用本文:

李建福罗程^* 董立尧德中^#. 复杂网络主成分分析的分类方法在音乐家白质可塑性研究中的应用[J]. 中国生物医学工程学报, 2015, 34(2): 184-189.
Li Jianfu Luo Cheng^* Dong Li Yao Dezhong^#. Classification of Principal Component Analysis on Complex Network and Application for White Matter Plasticity of Musicians. journal1, 2015, 34(2): 184-189.

链接本文:

http://cjbme.csbme.org/CN/10.3969/j.issn.0258-8021.2015.02.008 或 http://cjbme.csbme.org/CN/Y2015/V34/I2/184

［1］Sporns O, Chialvo DR, Kaiser M, et al. Organization, development and function of complex brain networks ［J］. Trends Cogn Sci, 2004, 8(9): 418-425.
［2］尧德中, 罗程, 雷旭, 等. 脑成像与脑连接［J］. 中国生物医学工程学报, 2011, 30(1): 6-10.
［3］Li Jianfu, Luo Cheng, Peng Yueheng, et al. Probabilistic diffusion tractography reveals improvement of structural network in musicians ［J］. PLoS ONE, 2014, 9(8): e105508.［4］蒋田仔, 刘勇, 李永辉. 脑网络：从脑结构到脑功能［J］. 生命科学, 2009, 21(2): 181-188.
［5］Jolliffe I. Principal component analysis ［M］. Hoboken：Wiley Online Library，2005.
［6］Munte TF, Altenmuller EL. The musician's brain as a model of neuroplasticity ［J］. Nature Reviews Neuroscience, 2002, 3(6): 473-477.
［7］Meister I, Krings T, Foltys H, et al. Effects of longterm practice and task complexity in musicians and nonmusicians performing simple and complex motor tasks: implications for cortical motor organization ［J］. Hum Brain Mapp, 2005, 25(3): 345-352.
［8］Huang Z, Zhang JX, Yang Z, et al. Verbal memory retrieval engages visual cortex in musicians ［J］. Neuroscience, 2010, 168(1): 179-189.
［9］Gaser C, Schlaug G. Brain structures differ between musicians and nonmusicians ［J］. J Neurosci, 2003, 23(27): 9240-9245.
［10］Keenan JP, Thangaraj V, Halpern AR, et al. Absolute pitch and planum temporale ［J］. NeuroImage, 2001, 14(6): 1402-1408.
［11］Schmithorst VJ, Wilke M. Differences in white matter architecture between musicians and nonmusicians: a diffusion tensor imaging study ［J］. Neuroscience Letters, 2002, 321(1): 57-60.
［12］Imfeld A, Oechslin MS, Meyer M, et al. White matter plasticity in the corticospinal tract of musicians: a diffusion tensor imaging study ［J］. NeuroImage, 2009, 46(3): 600-607.
［13］Leemans A, Jones DK. The Bmatrix must be rotated when correcting for subject motion in DTI data ［J］. Magn Reson Med, 2009, 61(6): 1336-1349.
［14］Tzourio-Mazoyer N, Landeau B, Papathanassiou D, et al. Automated anatomical labeling of activations in SPM using a macroscopic anatomical parcellation of the MNI MRI singlesubject brain ［J］. NeuroImage, 2002, 15(1): 273-289.
［15］Zalesky A, Fornito ABullmore ET. Networkbased statistic: identifying differences in brain networks ［J］. NeuroImage, 2010, 53(4): 1197-1207.
［16］Halpern AR, Zatorre RJ. When that tune runs through your head: a PET investigation of auditory imagery for familiar melodies ［J］. Cerebral Cortex, 1999, 9(7): 697-704.
［17］Tillmann B, Janata P, Bharucha JJ. Activation of the inferior frontal cortex in musical priming ［J］. Brain Res Cogn Brain Res, 2003, 16(2): 145-161.
［18］Sluming V, Barrick T, Howard M, et al. Voxelbased morphometry reveals increased gray matter density in Broca's area in male symphony orchestra musicians ［J］. NeuroImage, 2002, 17(3): 1613-1622.
［19］Blood AJ, Zatorre RJ. Intensely pleasurable responses to music correlate with activity in brain regions implicated in reward and emotion ［J］. Proc Natl Acad Sci USA, 2001, 98(20): 11818-11823.