Hi Daniel,
You don't say which version of SPM you are using, or what format your
images are in, so it's difficult to be sure. I've noticed that this
happens with my resize_img script under SPM5 when using 4D NIfTI
series. I think SPM is probably worried that NIfTI volumes can only
have one affine voxel-world mapping for the entire 4D series, so
instead, it saves a .mat file with one per 3D volume.
However, in the cases I've noticed with
http://www.cs.ucl.ac.uk/staff/gridgway/vbm/resize_img.m
all the matrices stored in the .mat (as M or mat, I forget which) just
feature the same set of scales and translations, without any rotation.
This should mean that other software will correctly interpret the
images without being able to read the .mats, and SPM (at least, SPM5
with NIfTI) should be perfectly happy if you delete the .mat files,
since they contain no information beyond what's in the NIfTI header.
Please reply with more information if this doesn't seem to be the
case. E.g. if you type:
whos -file image_blah.mat
load image_blah.mat
in Matlab, what variables can you see, and what values do these have?
Best,
Ged
> We have been attempting to reslice some realigned images. We would like
> to be able to end up with images that have no mat files. However, after
> running spm_reslice we still have mat files associated with the r*.img
> files. Is there something that we are doing wrong or another step we
> need to take?
>
> Thank you,
|