I figured out it has something to do with the DLH value
The data I'm performing these tests on is perfusion data from 14 participants before and after an physical exercise intervention.
The main problem is that the DLH value is too low (I think). If I run the cluster command with an artificial number 1 for example I do get p-values that make sense.
Which means I need to get a reasonable DLH value. So the question that remains is:
How can I calculate the correct DLH value from a perfusion image?
If I run the smoothest function on my res4d image with verbrose option turned on I get the following output:
zstatname = zstat
Reading mask....done
mask:: Size = (91,109,91,1)
mask:: Dims = (2,2,2,1)
mask:: ROI Size = (91,109,91,1)
mask:: Minimum and maximum intensities are: 0 and 1
Reading datafile....done
Data (residuals/zstat):: Size = (91,109,91,28)
Data (residuals/zstat):: Dims = (2,2,2,1)
Data (residuals/zstat):: ROI Size = (91,109,91,28)
Data (residuals/zstat):: Minimum and maximum intensities are: -88.863 and 88.863
Standardising....done
Masked-in voxels = 228483
Non-edge voxels = 215509
(v - 2)/(v - 1) = 1.03571
SSminus[X] = nan, SSminus[Y] = nan, SSminus[Z] = nan, S2[X] = nan, S2[Y] = nan, S2[Z] = nan
DLH nan voxels^-3 before correcting for temporal DOF
FWHMx = nan voxels, FWHMy = nan voxels, FWHMz = nan voxels
FWHMx = nan mm, FWHMy = nan mm, FWHMz = nan mm
DLH nan voxels^-3
VOLUME 228483 voxels
RESELS nan voxels per resel
DLH nan
VOLUME 228483
RESELS nan
FWHMvoxel nan nan nan
FWHMmm nan nan nan
########################################################################
To unsubscribe from the FSL list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1
|