Perhaps I am missing something very basic here but I am running into a problem that I don't understand. My EPI images need to be flipped in order to match the orientation of the EPI template. In the Display function I can do this by changing the pitch and roll. and apply it tothe images. However, when I go to 'Check Reg' to view the images, they appear back in their native orientation. Then, when I view the images in 'Display' they appear in the altered orientation (the orientation that matches the template) What is going on here?
>>> John Ashburner <[log in to unmask]> 12/13/01 07:23 AM >>>
> We are trying to coregister EPIs to a T1 anatomy in a patient with a
> tumorous brain lesion. To avoid distortion by the lesion we decided to mask
> most of it out. We then coregistered the altered EPIs to the T1 anatomy.
Editing the EPI images is probably not the best way to proceed. If the
default SPM99 coregistration is not working (because it assumes that brains
consist only of grey matter and white matter), then you may like to try
the updated mutual information coregistration routine that I mentioned on the
list recently.
>
> The next step hasn't worked so far: using the generated .mat-file to
> reslice the original, unaltered EPIs to the T1, and do the statistics with
> the resliced EPIs. I have renamed the .mat-file so it matches the EPI file
> name, but I still get error messages. Does anybody know an answer? Does
> anybody think this procedure makes no sense at all?
It depends what the error messages were. They usually contain some clues.
>
> Another thought: to improve (?) coregistration we would like to tell SPM
> that all zeros (after masking) are NaNs. I played around with matlab, but
> couldn't convince it to save the altereddata set...
You should be able to do this by directly calling spm_imcalc via the matlab
command line.
VI = spm_vol(spm_get);
VO = VI;
VO.fname = 'test.img';
VO.dim(4) = spm_type('float');
VO = rmfield(VO,'pinfo');
f = 'i1.*i1./i1'
VO = spm_imcalc(VI,VO,f)
I don't know if this is likely to help the coregistration though.
best regards,
-John
--
Dr John Ashburner.
Functional Imaging Lab., 12 Queen Square, London WC1N 3BG, UK.
tel: +44 (0)20 78337491 or +44 (0)20 78373611 x4381
fax: +44 (0)20 78131420 http://www.fil.ion.ucl.ac.uk/~john
|