Florey Institute of Neuroscience and Mental Health, Melbourne Brain Centre, 245 Burgundy Street, Heidelberg, VIC, 3084, Australia.
Florey Department of Neuroscience and Mental Health, University of Melbourne, Melbourne, VIC, Australia.
Brain Struct Funct. 2017 Nov;222(8):3761-3774. doi: 10.1007/s00429-017-1431-1. Epub 2017 Apr 26.
Interest in the study of brain connectivity is growing, particularly in understanding the dynamics of the structural/functional connectivity relation. Structural and functional connectivity are most often analysed independently of each other. Track-weighted functional connectivity (TW-FC) was recently proposed as a means to combine structural/functional connectivity information into a single image. We extend here TW-FC in two important ways: first, all the functional data are used without having to define a prior functional network (cf. TW-FC generates a map for a pre-specified network); second, we incorporate time-resolved connectivity information, thus allowing dynamic characterisation of functional connectivity. We refer to this technique as track-weighted dynamic functional connectivity (TW-dFC), which fuses structural/functional connectivity data into a four-dimensional image, providing a new approach to investigate dynamic connectivity. The structural connectivity information effectively 'constrains' the extremely large number of possible connections in the functional connectivity data (i.e. each voxel's connection to every other voxel), thus providing a way of reducing the problem's dimensionality while still maintaining key data features. The methodology is demonstrated in data from eight healthy subjects, and independent component analysis was subsequently applied to parcellate the corpus callosum, as an illustration of a possible application. TW-dFC maps demonstrate that different white matter pathways can have very different temporal characteristics, corresponding to correlated fluctuations in the grey matter regions they link. A realistic parcellation of the corpus callosum was generated, which was qualitatively similar to topography previously reported. TW-dFC, therefore, provides a complementary new tool to investigate the dynamic nature of brain connectivity.
人们对脑连接的研究兴趣日益浓厚，尤其是在理解结构/功能连接关系的动态方面。结构连接和功能连接通常是相互独立地进行分析的。最近提出了基于轨迹加权的功能连接（TW-FC），作为将结构/功能连接信息整合到单个图像中的一种方法。我们在此从两个重要方面扩展了 TW-FC：首先，无需定义先验功能网络即可使用所有功能数据（与 TW-FC 生成针对预定义网络的映射相比）；其次，我们整合了时变连接信息，从而能够对功能连接进行动态描述。我们将这种技术称为基于轨迹加权的动态功能连接（TW-dFC），它将结构/功能连接数据融合到一个四维图像中，为研究动态连接提供了一种新方法。结构连接信息有效地“约束”了功能连接数据中可能存在的大量连接（即每个体素与其他体素的连接），从而提供了一种降低问题维度的方法，同时仍保留关键数据特征。该方法在 8 位健康受试者的数据中得到了验证，随后应用独立成分分析对胼胝体进行分区，作为可能应用的一个示例。TW-dFC 图谱表明，不同的白质通路可能具有非常不同的时间特征，与它们连接的灰质区域的相关波动相对应。生成了胼胝体的现实分区，其与先前报道的地形学具有定性相似性。因此，TW-dFC 提供了一种新的补充工具，用于研究脑连接的动态性质。
