Abstract
Resting state networks (RSNs) have been found in human brains during awake resting states. RSNs are composed of spatially distributed regions in which spontaneous activity fluctuations are temporally and dynamically correlated. A new computational framework for reconstructing RSNs with human EEG data has been developed in the present study. The proposed framework utilizes independent component analysis (ICA) on short-time Fourier transformed inverse source maps imaged from EEG data and statistical correlation analysis to generate cortical tomography of electrophysiological RSNs. The proposed framework was evaluated on three sets of resting-state EEG data obtained in the comparison of two conditions: (1) healthy controls with eyes closed and eyes open; (2) healthy controls and individuals with a balance disorder; (3) individuals with a balance disorder before and after receiving repetitive transcranial magnetic stimulation (rTMS) treatment. In these analyses, the same group of five RSNs with similar spatial and spectral patterns were successfully reconstructed by the proposed framework from each individual EEG dataset. These EEG RSN tomographic maps showed significant similarity with RSN templates derived from functional magnetic resonance imaging (fMRI). Furthermore, significant spatial and spectral differences of RSNs among compared conditions were observed in tomographic maps as well as their spectra, which were consistent with findings reported in the literature. Beyond the success of reconstructing EEG RSNs spatially on the cortical surface as in fMRI studies, this novel approach defines RSNs further with spectra, providing a new dimension in understanding and probing basic neural mechanisms of RSNs. The findings in patients' data further demonstrate its potential in identifying biomarkers for the diagnosis and treatment evaluation of neuropsychiatric disorders.
Highlights
Over the past decade, research on the human brain has been increasingly drawn toward investigation of networked brain activity among different brain regions during resting states, termed as the resting state networks (RSNs) (Biswal et al, 1995; Fox and Raichle, 2007)
Resting state functional magnetic resonance imaging (fMRI) studies have identified various RSNs associated with different brain functions (De Luca et al, 2006), demonstrated their consistency in healthy subjects (Beckmann et al, 2005; Damoiseaux et al, 2006), alterations in neuropsychiatric disorders (Rombouts et al, 2005; Greicius et al, 2007; Sorg et al, 2007; Agosta et al, 2012), and changes with cognitive tasks (Greicius et al, 2003; Buckner et al, 2008)
Ni′ (j, k) − 3 · zi Ni′ (j, k) − 3, i = 1, . . . N and Cluster-Based Statistical Thresholding To quantitatively define brain regions that belong to an RSN, statistical correlation coefficient (SCC) maps obtained after Equation (11) need to be thresholded
Summary
Research on the human brain has been increasingly drawn toward investigation of networked brain activity among different brain regions during resting states, termed as the resting state networks (RSNs) (Biswal et al, 1995; Fox and Raichle, 2007). Among various methods for probing resting-state brain signals, independent component analysis (ICA) has been widely used to identify RSNs from both EEG/MEG and fMRI data. It is noted that these techniques make it feasible to directly compare fMRI RSNs and EEG/MEG RSNs in a common spatial domain, providing new insights to the electrophysiological basis and hemodynamic aspects of RSNs, especially when fMRI and EEG data can be simultaneously recorded (Goldman et al, 2002; Gonçalves et al, 2006; Yuan et al, 2016). Despite these new advancements, methods to derive EEG/MEG RSNs are still limited in many ways. There were two conditions (i.e., eyes-open vs. eyes-closed; healthy individuals vs. patients; and before vs. after treatment) and their comparisons were used to examine the resolution and capability of the proposed framework in identifying condition-specific differences
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
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 call
