All children with concussion were scanned at a single site (Imaging Research Centre [IRC] at St. Joseph’s Healthcare, Hamilton) using a 3-Tesla GE Discovery MR750 MRI scanner and a 32-channel phased array head receiver coil. Upon entering the IRC, participants (as well as their parents and/or guardians if aged 16 years or younger) were led through an intake questionnaire by the MRI technologist, who then situated the participant in the MRI, using foam cushioning to minimize discomfort and motion during the scan. The MRI technologist remained in verbal contact with the participant via intercom throughout the scan.
With respect to MRI data collection, first, a 3-plane localizer sequences was acquired. Anatomical images were then collected using a 3D inversion recovery-prepped fast SPGR T1-weighted sequence (TR/TE = 11.36/4.25 ms, TI = 450mms, flip angle = 12°, 512 × 256 matrix interpolated to 512 × 512, 22 cm axial FOV, 1 mm thick). Resting state fMRI (rs-fMRI) involved BOLD imaging (gradient echo EPI, TR/TE = 2000/35 ms, flip angle = 90°, 64 × 64 matrix, 180 time points, 3 mm thick, 22 cm FOV), wherein participants were asked to remain awake, keep their eyes open, and not to think of anything in particular. A B0 map was acquired for resting state scans, using the same geometric prescription as the rs-fMRI scan. A Bo mapping tool available on the GE scanner provided a parametric map of field homogeneity in Hz. In regards to the scanning sequence, the rs-fMRI data were acquired within 10-min of entering the MRI, as to avoid motion onset by restlessness later in scans as we have observed in this population. Additional data were collected (including DTI35 (link) and task-based fMRI data36 (link)) as part of the imaging battery, but are not relevant to the present study.
With respect to control data from the ABIDE-II database, only scans with a minimum of 180 time-points were used as age- and sex-matched controls. B0 data were not available for healthy controls. Approximately 10% of the ABIDE-II data did not meet our quality control processes (as implemented in CONN 21a, as below) and were ultimately discarded from the analysis.
With respect to MRI data collection, first, a 3-plane localizer sequences was acquired. Anatomical images were then collected using a 3D inversion recovery-prepped fast SPGR T1-weighted sequence (TR/TE = 11.36/4.25 ms, TI = 450mms, flip angle = 12°, 512 × 256 matrix interpolated to 512 × 512, 22 cm axial FOV, 1 mm thick). Resting state fMRI (rs-fMRI) involved BOLD imaging (gradient echo EPI, TR/TE = 2000/35 ms, flip angle = 90°, 64 × 64 matrix, 180 time points, 3 mm thick, 22 cm FOV), wherein participants were asked to remain awake, keep their eyes open, and not to think of anything in particular. A B0 map was acquired for resting state scans, using the same geometric prescription as the rs-fMRI scan. A Bo mapping tool available on the GE scanner provided a parametric map of field homogeneity in Hz. In regards to the scanning sequence, the rs-fMRI data were acquired within 10-min of entering the MRI, as to avoid motion onset by restlessness later in scans as we have observed in this population. Additional data were collected (including DTI35 (link) and task-based fMRI data36 (link)) as part of the imaging battery, but are not relevant to the present study.
With respect to control data from the ABIDE-II database, only scans with a minimum of 180 time-points were used as age- and sex-matched controls. B0 data were not available for healthy controls. Approximately 10% of the ABIDE-II data did not meet our quality control processes (as implemented in CONN 21a, as below) and were ultimately discarded from the analysis.
Free full text: Click here
