Files are available for download in Analyze format. The13th template is the binary combination of the left and right executive networks.
2. Left Executive
3. Right Executive
8. Anterior Cingulate/PreCun
9. Parietal Association Cortex
10. Supplementary Motor
11. Posterior Default
12. IFG/Middle Temporal
13. Left-Right Executive Combined Network
How Templates Were Made
On average 6-minutes of eyes-closed, awake resting state data was collected for 62 children ages 9-15. During the resting-state scan, 29 axial slices of blood-oxygen-level-dependent (BOLD) data were acquired with 4-mm slice thickness (no skip). A T2*-sensitive gradient echo spiral in/out pulse sequence was used for BOLD data acquisition (TR = 2000ms, TE = 30ms, flip angle = 77°, FOV = 22 cm, 64 x 64). Data were collected at the Richard M. Lucas Center for Imaging at Stanford University. Processing steps and template development steps are described below. Rest-retest reliability of these networks has recently been demonstrated within children, some of whom contributed to this data set (Thomason et al., 2010).
1. Physiological correction in data reconstruction
Resting-state fMRI images were preprocessed using a correction that diminishes BOLD signal fluctuations contributed by respiratory and HR variations. Using the method developed by Chang and Glover (2009), this correction reduces the effect of low-frequency respiratory variations (i.e., the “envelope” of the respiratory belt waveform) and HR (average rate in a 6-sec sliding window) by first convolving those independently measured signals with appropriate filters and then regressing them out of the time series for each voxel (Chang and Glover, 2009).
2. fMRI preprocessing
Participant data were preprocessed using Statistical Parametric Mapping software (SPM8). Preprocessing included image realignment and co-registration of functional and anatomical images. Functional images were normalized to the Montreal Neurological Institute (MNI) template, using the participant-specific transformation parameters created by fitting mean functional images to the single reference EPI standard SPM template. Following normalization, all participant images were visually inspected, and we determined that all normalizations had proceeded satisfactorily. Images were smoothed with a 6-mm Gaussian kernel to decrease spatial noise.
3. Movement correction
Movement was plotted and visually inspected for every participant. Time frames corresponding to brief movement spikes (> 0.8mm) and lasting less than 5 frames were removed. This correction resulted in low average movement across participants (< 0.5 mm). In addition, the first three time frames were removed for all participants to allow for signal stabilization. In total, no more than 10% of time frames were removed for a given participant. Remaining time frames for all participants were submitted to group ICA; these are reported for each participant in the linked xls file.
4. Group ICA analysis
Remaining time frames for all 62 participants were concatenated in a group independent component analysis (ICA) implemented in Matlab using GIFT (Calhoun et al., 2001). After an automated template-matching algorithm was used to determine those individual components corresponding to the default-mode, left and right executive, salience, motor, auditory, and visual resting-state networks, single-participant spatial maps were back reconstructed for use in group random-effects analysis that follows.
5. Template generation
SPM8 was used to perform one sample t-tests (separately for each network) on participant ICA maps. Resulting maps were threshold at p05 K min 25, FWE-corrected. These were saved & smoothed at 8mm in SPM8. Using the Imcalc subroutine in SPM, the smoothed results images were then converted to binary mask images using the equation i1>1 (performed separately for each mask). Finally, the binarized, smoothed images were resampled to the resolution 3.44 x 3.44 x 4 using the co-register function in SPM8.
The work was performed by Drs. Moriah E. Thomason, Ian H. Gotlib & Gary H. Glover during a collaborative developmental study at Stanford University. Support for this project was provided by F32-MH081583 to MET, P41-RR009784 to GHG, RO1-MH074849 to IHG, and a NARSAD young investigator award to MET.
1. Calhoun VD, Adali T, Pearlson GD, & Pekar JJ. (2001). A method for making group inferences from functional MRI data using independent component analysis. Hum Brain Mapp. 14:140-151.
2. Chang C, & Glover GH. (2009). Effects of model-based physiological noise correction on default mode network anti-correlations and correlations. Neuroimage. 47:1448-1459.
3. Thomason ME, Dennis EL, Joshi AA, Joshi SH, Dinov ID, et al. (on line, 12/5/2010). Resting-state fMRI can reliably map neural networks in children. Neuroimage.
The author distributes these software materials with no guarantees explicit or implied, and neither she nor sponsoring institutions accepts liability for any use of these templates