comqa - Simple fMRI stability test
Command line program to analyse dynamic series of a phantom, find and track any
changes in the position of the centre of mass of the phantom (assuming the
phantom is spanned
in all three dimensions) and calculate mean and standard deviation of
intensities values in a spherical region of interest centred on the COM.
Results are written to both text and CSV files. COM values are in units
of voxels (be
aware of this if the input Analyze file has non-isotropic voxels, resulting in
a non-spherical ROI in real units).
comqa [-d] [-rad n] [-ct] [-tsnr] [-acr] [Analyze/NIFTI filename] [-p phase image filename]
If no command line options are entered, the program just requests the
multivolume input Analyze filename. Optional command line flags may be
specified -
- -d
-
Outputs extra diagnostic information for debugging.
- -rad n
-
Overrides the default radius of the spherical ROI by n pixels. If
n is negative, then set the radius automatically (currently 10% of the
matrix size).
Whether this flag is specified or not, the value of the radius used is
output, in pixels.
- -ct
-
Assumes the input images are from a CT scan rather than MRI. In the MR images,
the edge of the object is the liquid/plastic boundary, whereas in CT images
both the liquid and the plastic are visualised, which means different
phantom diameters will be measured depending on modality. This option
thresholds the CT image to around -20 HU so that the FWHM used to calculate
the phantom diameters should find the water/plastic boundary rather than the
plastic/air boundary.
Any objects with similar intensities (e.g., couch, or head holder) may
need to be removed from the CT images prior to running this program.
- -tsnr
-
Also calculate and save temporal SNR images.
- -acr
-
Assuming the images are of the large ACR phantom, performs some basic
calculations on slice 7 of 5.Image Intensity Uniformity and
6.Percent Signal Ghosting. The other values in the results file
should not be used.
-p
Applies the same ROI to phase images of the same dimensions as the modulus
images, and calculate the mean phase and the standard deviation within the ROI.
If the scaling and offset parameters are set in the input phase image header,
then these will be applied to the phase values.