Date: 03-13-2018
Link: http://rpubs.com/navona/SPINS_PRISMA

This script compares data from two pilot participants scanned on the new ToNI CMP scanner to 10 randomly selected MRP and ZHP participants, on metrics central to structural, functional, and diffusion imaging. (Note: there are fewer than 10 MRP and ZHP participants included in the diffusion analysis, as some had not made it fully through the DTI pipelines.)

The scatterplots provide a visual overview of the data distribution. All values have been Z-transformed. The dotted line indicates 1.5 SDs away from mean, and the solid line indicates 2.0 SDs away from mean.

The associated tables provide definitions for all metrics, and an indication of if the pilot participants appear to be different from the MRP and ZHP participants: ‘1’ indicates that 1 CMP pilot fell >= 2.0 SDs away from mean, and ‘2’ indicates that both CMP pilots did so.

Structural

source: T1-w
QC pipeline: MRIQC
citation: https://mriqc.readthedocs.io/en/latest/iqms/t1w.html

metric definition count
Measures based on noise measurements
cjv Coefficient of joint variation. High values relate to heavy head motion and large INU artifacts. 0
cnr Contrast-to-noise ratio. Evaluates how separated tissue distributions of GM and WM are. Higher values indicate better quality. 0
snr_csf Signal-to-noise ratio, in CSF. Calculated within the tissue mask. 0
snr_gm Signal-to-noise ratio, in GM. Calculated within the tissue mask. 0
snr_wm Signal-to-noise ratio, in WM. Calculated within the tissue mask. 0
snr_total Signal-to-noise ratio, across CSF, GM, and WM. Calculated within the tissue mask. 0
snrd_csf Dietrich’s SNF, in CSF. Uses the air background as reference. 1
snrd_gm Dietrich’s SNF, in GM. Uses the air background as reference. 1
snrd_wm Dietrich’s SNF, in WM. Uses the air background as reference. 0
snrd_total Dietrich’s SNF, across CSF, GM, and WM. Uses the air background as reference. 0
qi_2 Mortamet’s quality index. Calculation of goodness-of-fit of a chi-square distribution on the air mask. 0
Measures based on information theory
efc Entropy-focus criterion. Indication of ghosting and blurring induced by head motion. Lower values are better. 2
fber Foreground-background energy ratio. Mean energy of image values within the head relative to outside the head. Higher values are better. 2
Measures targeting specific artifacts
inu_med Median of the INU field. Values closed to 1.0 are better. 0
inu_range Range of the INU field. Values closed to 1.0 are better. 1
qi_1 Proportion of voxels with intensity corrupted by artifacts. Lower values are better. 0
wm2max White-matter to maximum intensity ratio. Values should be around the interval [0.6, 0.8]. 0
Other measures
fwhm_avg Spatial distribution of the image intensity values in units of voxels. Lower values are better. 0
rpve_csf Residual partial volume effect, of CSF. Lower values are better. 0
rpve_gm Residual partial volume effect, of GM. Lower values are better. 0
icvs_csf Volume fractions, of CSF. Should move within a normative range. 0
icvs_gm Volume fractions, of GM. Should move within a normative range. 0
icvs_wm Volume fractions, of WM. Should move within a normative range. 0

Diffusion

source: DTI 60-1000
QC pipeline: QCMON
citation: https://github.com/josephdviviano/qcmon

metric definition count
maxRELrms Maximum relative root mean square of displacement. Measure of head motion. Should be low. 0
maxABSrms Maximum RMS displacement relative to a single reference volume. Measure of head motion. Should be low. 0
meanRELrms Average relative root mean square of displacement. Measure of head motion. Should be low. 0
meanABSrms Average RMS displacement relative to a single reference volume. Measure of head motion. Should be low. 0
outcount_bX Count of number of temporal outliers. 0
outmean_bX Mean value of temporal outliers, across all directions. 0
outmax_bX Max value of temporal outliers, across all directions. 0
driftpercent Percentage of drift of global signal, across all directions. 0
drift_bX Drift of global signal mean, across all directions. 0
gmean_bX Global signal mean, across all directions. 1
tsnr_bX Average temporal SNR, across all directions. Should be high. 0
clipcount Number of voxels with clipped amplitude (i.e. >= 4095). 2

Functional

QC pipeline: MRIQC
source: Empathic Accuracy (EA)

source: Imitate task (IMI)

source: Observe task (OBS)

source: Resting state

metric definition EA IMI OBS RS
Measures for the structural information
efc Entropy-focus criterion. Indication of ghosting and blurring induced by head motion. Lower values are better. 0 0 0 0
fber Foreground-background energy ratio. Mean energy of image values within the head relative to outside the head. Higher values are better. 0 0 0 0
fwhm_avg Average full-width half maximum smoothness. 0 0 0 0
bg_median Median of background SNR. 1 0 0 0
fg_median Median of foreground SNR. 1 1 1 1
Measures for the temporal information
dvars_nstd Indexes the rate of change of BOLD signal across the entire brain at each frame of data. 0 0 0 0
dvars_std As dvars_nstd, but normalized with the standard deviation of the temporal difference time series. 0 0 0 0
dvars_vstd As dvars_nstd, but with voxel-wise standardization. 0 0 0 0
gcor Calculates an optimized summary of time-series. 0 0 0 0
Measures for artifacts and other
tsnr Temporal SNR. Should be high. 0 0 0 0
fd_mean Expresses instantaneous head motion. 0 0 0 0
fd_num Number of timepoints above FD threshold (.2mm). 0 0 0 0
fd_perc Percent of timepoints above FD threshold (.2mm). 0 0 0 0
gsr_x Ghost to signal ratio, along x phase-encoding axis. 0 0 0 0
gsr_y Ghost to signal ratio, along y phase-encoding axis. 0 0 0 0
aor Mean fraction of outliers per fMRI volume. 0 0 0 0
aqi Mean quality index. 0 0 0 0