Dear SPM list,
In event-related fMRI processing I should set up the "Number of
Bins/TR" and "Sample bin" according to what I have used at the Slice
Timing correction step (the very 1st one).
However, if I perform spatial normalization with re-slicing during
prestatistic processing, my original slice orientation, slice
thickness, and the number of slices change.
How can I be sure that I choose the sample bin correctly in this case?
In my case, I have fMRI data as a series of 20 axial slices (bottom to
top, first odd, then even numbers). TR=2sec. Slice thickness= 4.4 mm
First, I corrected the data to the 6th slice (which is 13th in time),
because it crosses an area of interest. Then, after realignment (with
no reslicing), I perform the spatial normalization with reslicing,
changing Z voxel size form 4.4 down to 3 mm (in order to be able to
use smaller spatial filter later) and giving me 46 slices instead of
former 20.
Moreover, since the original slice orientation was more or less
parallel to STS, the slice orientation in normalized data has changed
pretty much from the original one.
Finally, at the statistical step I need to specify the "Number of
Bins/TR" and "Sample bin" so that the model would be synchronized with
the same slice that has been chosen at the slice timing correction
step, right? ... and how do I do that if my images have been
reoriented and resliced?
thanks a lot in advance!
Victor
---------------------
Victor Vorobyev, Ph.D.
Centre for Cognitive Neuroscience
Department of Psychology
University of Turku
Finland
|