Functional connectivity in resting state fMRI data: A small-world wavelet correlation approach
-
1
University of Geneva, Switzerland
-
2
Swiss Federal Institute of Technology, Switzerland
Two main approaches have been proposed to characterize patterns of functional connectivity across brain regions during resting state conditions: (i) independent component analysis (ICA) and (ii) ROI-based General Linear Model (GLM) methods. These techniques have been used to identify brain regions that are synchronously active during wakeful resting state and deactivated during cognitive tasks and that constitute the “default mode” network. ICA produces a collection of maps showing regions that tend to share similar fMRI signal change across time. Exploratory ICA maps can be generated without any a priori constraints about the patterns of functional connectivity. However, ICA has important drawbacks because the selection of meaningful maps is typically performed using empirical prior knowledge, and because the statistical assessment of individual maps remains largely tentative. Traditional ROI-based GLM approach in fMRI starts with the selection of a presumably significant component of a network of interest. The time-series of activity extracted from this component is then entered into a GLM to assess its correlation with the time-series of any other region in the brain. There are important limitations of a GLM approach to design-free data because it requires an a priori identification of seed regions, and because it cannot readily identify several simultaneous and partially overlapping networks. Critically, when correlating activity from one seed region with the rest of the brain, the conventional statistical assessment for GLM-based results cannot be applied due to the large amount of background correlation that is present in fMRI data. Here we evaluate a new methodological framework that is well suited for the analysis of functional connectivity across brain regions and that improves the determination of significance levels across functional connectivity networks. We first apply a “small-world wavelet correlation” (SWWC) approach to the fMRI time-series to assess functional brain connections during resting state. For each subject, we build a personal, anatomical atlas delineating 90 brain regions and compute cross-correlations between all these segmented regions. To do so, we use discrete wavelet transform to generate cross-correlation matrices for different temporal frequency bands. From these matrices we create 3-D connectivity maps based on a false discovery rate thresholding. This approach also allows us to analyze the temporal behavior of connectivity networks by using sliding time-windows, producing correlation matrices and connectivity maps for each time point. In this way, we can study the dynamic features of functional connectivity networks, and possible changes across experimental conditions. Finally, for each thresholded cross-correlation matrix we can compute the node statistics in order to evaluate the characteristics of the network and in particular its similarity to a small-world network, i.e., a network that uses a limited number of connections to efficiently link the nodes in the network. In summary, the SWWC approach involves a parcellization of the whole brain into functional regions, flexible temporal assessment of multiple, simultaneous networks, and robust statistical testing for the presence of connectivity over time. We suggest that this approach may offer major advantages over traditional methods for the study of temporal functional coupling of design-free fMRI data.
Conference:
10th International Conference on Cognitive Neuroscience, Bodrum, Türkiye, 1 Sep - 5 Sep, 2008.
Presentation Type:
Poster Presentation
Topic:
Neuroinformatics of Cognition
Citation:
Eryilmaz
H,
Van De Ville
D,
Schwartz S
S and
Vuilleumier
P
(2008). Functional connectivity in resting state fMRI data: A small-world wavelet correlation approach.
Conference Abstract:
10th International Conference on Cognitive Neuroscience.
doi: 10.3389/conf.neuro.09.2009.01.350
Copyright:
The abstracts in this collection have not been subject to any Frontiers peer review or checks, and are not endorsed by Frontiers.
They are made available through the Frontiers publishing platform as a service to conference organizers and presenters.
The copyright in the individual abstracts is owned by the author of each abstract or his/her employer unless otherwise stated.
Each abstract, as well as the collection of abstracts, are published under a Creative Commons CC-BY 4.0 (attribution) licence (https://creativecommons.org/licenses/by/4.0/) and may thus be reproduced, translated, adapted and be the subject of derivative works provided the authors and Frontiers are attributed.
For Frontiers’ terms and conditions please see https://www.frontiersin.org/legal/terms-and-conditions.
Received:
15 Dec 2008;
Published Online:
15 Dec 2008.
*
Correspondence:
Hamdi Eryilmaz, University of Geneva, Lausanne 1015 Vaud, Switzerland, hamdi.eryilmaz@gmail.com