A method for independent component graph analysis of resting-state fMRI

Abstract

Introduction: Independent component analysis (ICA) has been extensively used for reducing task-free BOLD fMRI recordings into spatial maps and their associated time-courses. The spatially identified independent components can be considered as intrinsic connectivity networks (ICNs) of non-contiguous regions. To date, the spatial patterns of the networks have been analyzed with techniques developed for volumetric data.

Objective: Here, we detail a graph building technique that allows these ICNs to be analyzed with graph theory.

Methods: First, ICA was performed at the single-subject level in 15 healthy volunteers using a 3T MRI scanner. The identification of nine networks was performed by a multiple-template matching procedure and a subsequent component classification based on the network "neuronal" properties. Second, for each of the identified networks, the nodes were defined as 1,015 anatomically parcellated regions. Third, between-node functional connectivity was established by building edge weights for each networks. Group-level graph analysis was finally performed for each network and compared to the classical network.

Results: Network graph comparison between the classically constructed network and the nine networks showed significant differences in the auditory and visual medial networks with regard to the average degree and the number of edges, while the visual lateral network showed a significant difference in the small-worldness.

Conclusions: This novel approach permits us to take advantage of the well-recognized power of ICA in BOLD signal decomposition and, at the same time, to make use of well-established graph measures to evaluate connectivity differences. Moreover, by providing a graph for each separate network, it can offer the possibility to extract graph measures in a specific way for each network. This increased specificity could be relevant for studying pathological brain activity or altered states of consciousness as induced by anesthesia or sleep, where specific networks are known to be altered in different strength.

Keywords: BOLD fMRI; graph theory; independent component analysis; resting state.

Figure 1

Processing workflow for functional networks: (a) 1, preprocessing: registration to T1 (SPM8); 2, independent component analysis (GIFT) and 3, neuronality check and template matching. (b) 1, Spatial normalization of T1 (SPM8), 2, Atlas‐based segmentation (FreeSurfer) and 3, parcellation (Nipype/CMTK, Lausanne 1,015 atlas). (c) 1, Extract regional fMRI timecourses (Y), 2, least squares solution to estimate β values using as predictors the timecourse from ICA as in (a) and 3, transform β values to t values. (d) 1, Calculate edge weights and draw networks for each independent component and 2, threshold correlation and anti‐correlation networks and analyze separately

Figure 2

Sagittal and axial representation of the network and cortical extent of the relative IC for the DMN. The degree of each one of the 1,015 regions is represented by the node's size and orange to yellow gradient. On the right, cortical parcellation for DMN extracted by group‐ICA superimposed to a structural T1 image

Figure 3

Graphical representation of the network and cortical extent of the relative ICs for the left and right executive control networks. The degree of each one of the 1,015 regions is represented by the node's size and color gradient. On the right column, cortical location for each network was extracted by group‐ICA superimposed to a structural T1 image

Figure 4

Graphical representation of the network and cortical extent of the relative ICs for auditory and salience networks. The degree of each one of the 1,015 regions is represented by the node's size and color gradient. On the right column, cortical location for each network was extracted by group‐ICA superimposed to a structural T1 image

Figure 5

Graphical representation of the network and cortical extent of the relative ICs for sensorimotor and visual lateral networks. The degree of each one of the 1,015 regions is represented by the node's size and color gradient. On the right column, cortical location for each network was extracted by group‐ICA superimposed to a structural T1 image

Figure 6

Graphical representation of the network and cortical extent of the relative ICs for visual medial and visual occipital networks. The degree of each one of the 1,015 regions is represented by the node's size and color gradient. On the right column, cortical location for each network was extracted by group‐ICA superimposed to structural T1 image

Figure 7

Weighted network pairs and significant difference analysis using post hoc paired comparison for each network property. The colors red and orange represent p‐values ≤.05 and ≤.01, respectively. Networks: X_CN (X represents the mask used to keep just the nodes belonging to the respective network and CN is the classical network), DMN (default mode network), AUD (auditory), ECL (executive control left), ECR (executive control right), SA (salience), SM (sensorimotor), VL (visual lateral), VM (visual medial) and VO (visual occipital)

Figure 8

Sagittal and axial representation of all nine networks (default mode network, executive control left, executive control right, visual lateral, visual medial, visual occipital, auditory, sensorimotor and salience) overlapped and classical network, with threshold 0.45. The average degree of each node is represented by the node's size and color gradient