Abstract
Detecting changes of spatially high-resolution functional connectivity patterns in the brain is crucial for improving the fundamental understanding of brain function in both health and disease, yet still poses one of the biggest challenges in computational neuroscience. Currently, classical multivariate Granger Causality analyses of directed interactions between single process components in coupled systems are commonly restricted to spatially low- dimensional data, which requires a pre-selection or aggregation of time series as a preprocessing step. In this paper we propose a new fully multivariate Granger Causality approach with embedded dimension reduction that makes it possible to obtain a representation of functional connectivity for spatially high-dimensional data. The resulting functional connectivity networks may consist of several thousand vertices and thus contain more detailed information compared to connectivity networks obtained from approaches based on particular regions of interest. Our large scale Granger Causality approach is applied to synthetic and resting state fMRI data with a focus on how well network community structure, which represents a functional segmentation of the network, is preserved. It is demonstrated that a number of different community detection algorithms, which utilize a variety of algorithmic strategies and exploit topological features differently, reveal meaningful information on the underlying network module structure.
Highlights
A comprehensive insight into brain processes requires an understanding of information flow between and within structures of the underlying neural system
We investigate the impact of the edge pattern information loss caused by dimension reduction on the quality and the recoverability of network modules in the resulting lsGCI networks
In addition to ROC curve analysis we considered the effect of different degrees of dimension reduction on edge pattern alterations and the recoverability of network modules in dichotomized lsGCI networks
Summary
A comprehensive insight into brain processes requires an understanding of information flow between and within structures of the underlying neural system. We propose a novel methodological concept, where spatially high-dimensional data are incorporated into connectivity analysis These data originate from a (possibly large) system of connected and interacting elements [8] and this system may be considered as a network [9], which represents the functional connectivity structure by linking a set of vertices (recording sites) by edges (interactions). We propose a general large scale Granger Causality (lsGC) approach, a purely data-driven procedure which involves incorporating a PCA data dimension reduction step into low-dimensional (LD) space, but attains connectivity patterns in the original high-dimensional (HD) space This concept is comprised of an orthogonal back projection of LD MVAR model residuals to HD space and using these back-transferred residuals for proper definitions of vertex by vertex interactions. All preprocessing steps were carried out using FEAT (FMRI Expert Analysis Tool), which is part of FSL and its respective subroutines
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
