Figure 2
Resting-state functional connectivity analysis pipeline. Preprocessing included skull stripping, co-registration to structural images, realignment to middle volume, spatial smoothing (6 mm FWHM), ICA-AROMA, high-pass filtering (0.008 Hz), nuisance regression, and MNI152-space warping (a). Mean activation time series for parcels of functional connectivity template20,21 were extracted after exclusion of “outside-field of view” parcels (90% of the time 80% of voxels in a parcel had to have an intensity of > 1000) (b, c). Individual Fisher’s z-transformed correlation matrices were created (d). Each correlation matrix was binarized using 7 different density thresholds (0.10, 0.15, 0.20, 0.25, 0.30, 0.35, 0.40) (e). Graphs were created based on binarized matrices and for 70 relevant nodes. Density-integrated as well as threshold-specific topological network measures were calculated (f, g). Figure was created using FSLeyes (version 0.22.6, https://fsl.fmrib.ox.ac.uk/fsl/fslwiki/FSLeyes) and R software (version 3.6.0, https://cran.r-project.org/).
