Brain functional and structural metrics

To compute the amplitude of low-frequency fluctuations based on resting-state fMRI data, the time series of a given voxel was extracted, and the fast Fourier transforms were performed to transform the time series into frequency domains. ALFF was calculated by averaging the square root of the power spectrum across 0.01–0.08 Hz. After calculating the ALFF values, the results were Z-scores and regressed out the demographic information (e.g., age, gender, and education years) for statistical analysis.

Voxel-based morphometry (VBM) analyses were performed to obtain whole-brain gray matter volume in both FS and healthy participants using T1-weighted images. The high-resolution T1-weighted images were processed with the Computational Anatomy Toolbox (CAT12) and SPM12 with the following preprocessing steps: (1) The T1 images were first manually set to the anterior commissure orientation for better registration after screening for artifacts and gross anatomical abnormalities; (2) the reorientated images were segmented into the gray matter (GM), white matter (WM), and cerebrospinal fluid (CSF) after signal inhomogeneities correction; (3) DARTEL was used with non-linear registration to standard space, and intensity modulation was conducted using the linear and non-linear components of the Jacobian determinant derived from the deformation fields; and (4) the intracranial volume was calculated by summing the volumes of the segmented gray matter, white matter, and cerebrospinal fluid compartments.

After performing the between-group comparison of the ALFF and GMV for identifying significant regions, a follow-up functional connectivity (FC) analysis was conducted on three ROIs. First, the resultant cluster of VBM analysis was first downsampled and registered to functional space (61 × 73 × 61). Subsequently, the overlap between significant clusters of structural and functional analysis was used as a region of interest for functional connectivity analysis. The mean time series across all voxels in the ROIs were extracted, and the voxel-wise functional connectivity was calculated. The resultant ROI-to-whole-brain FC maps were then Z-scored for further statistical analyses. The Data Processing Assistant for rs-fMRI (DPARSF; http://www.restfmri.net/forum/DPARSF) toolbox was used for FC analysis.