Original Research ARTICLE
The impact of the Geometric Correction Scheme on MEG functional topology at rest
- 1Università degli Studi G. d'Annunzio Chieti e Pescara, Italy
- 2Institute for Advanced Biomedical Technologies, G. d'Annunzio University of Chieti and Pescara, Italy
- 3Department of Neuroscience, School of Medicine and Surgery, University of Padova, Italy
- 4Padova Neuroscience Center, University of Padua, Italy
- 5Department of Neurology, Radiology, and Anatomy and Neurobiology, Washington University in St. Louis, United States
- 6ULB Neuroscience Institute, Free University of Brussels, Belgium
- 7Erasmus Hospital, Free University of Brussels, Belgium
- 8Faculty of Veterinary Medicine, University of Teramo, Italy
Spontaneous activity is correlated across brain regions in large scale networks (RSN) closely resembling those recruited during several behavioral tasks and characterized by functional specialization and dynamic integration. Specifically, MEG studies revealed a set of central regions (dynamic core) possibly facilitating communication among differently specialized brain systems. However, source projected MEG signals, due to the fundamentally ill-posed inverse problem, are affected by spatial leakage, leading to the estimation of spurious, blurred connections that may affect the topological properties of brain networks and their integration. To reduce leakage effects, several correction schemes have been proposed including the Geometric Correction Scheme (GCS) whose theory, simulations and empirical results on topography of a few RSNs were already presented. However, its impact on the estimation of fundamental graph measures used to describe the architecture of interactions among brain regions has not been investigated yet.
Here, we estimated dense, MEG band-limited power connectomes in theta, alpha, beta and gamma bands from 13 healthy subjects (all young adults). We compared the connectivity and topology of MEG uncorrected and GCS-corrected connectomes. The use of GCS considerably reorganized the topology of connectivity, reducing the local, within-hemisphere interactions mainly in the beta and gamma bands and increasing across-hemisphere interactions mainly in the alpha and beta bands. Moreover, the number of hubs decreased in the alpha and beta bands, but the centrality of some fundamental regions such as the Posterior Cingulate Cortex (PCC), Supplementary Motor Area (SMA) and Middle Prefrontal Cortex (MPFC) remained strong in all bands, associated to an increase of the Global Efficiency and a decrease of Modularity. As a comparison, we applied orthogonalization on connectomes and ran the same topological analyses. The correlation values were considerably reduced, and orthogonalization mainly decreased local within-hemisphere interactions in all bands, similarly to GCS. Notably, the centrality of the PCC, SMA and MPFC was preserved in all bands, as for GCS, together with other hubs in the posterior parietal regions. Overall, leakage correction removes spurious local connections, but confirms the role of dynamic hub regions, specifically the anterior and posterior cingulate, in integrating information in the brain at rest.
Keywords: Functional Connectivity, Band Limited Power correlation, MEG connectome, Leakage correction, Functional hubs
Received: 30 Apr 2019;
Accepted: 02 Oct 2019.
Copyright: © 2019 Della Penna, Corbetta, Wens and de Pasquale. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
* Correspondence: Prof. Stefania Della Penna, Università degli Studi G. d'Annunzio Chieti e Pescara, Chieti, Italy, firstname.lastname@example.org