Thanks again for your help and sorry for all the emails. The flow fields that 'Normalise to MNI' asks for are the individual subjects' flow fields 'u_c1*.nii which look like the 3rd image selected in the attached picture. Also included in this picture are a template, and subject's functional scan after the 'normalise to MNI' step.
The flow field looks pretty OK to me. I thought that you were
referring to something that was uniformly grey. This is the x
component. The other components can be viewed by looking at
u_rc1*_Template.nii,2 and u_rc1*_Template.nii,3 (by changing the 1 in
the file selector to 1:3).
Also note that the main objective of Dartel is to align all the scans
in the study together. When you normalise to MNI space, the images
that are closely aligned within study are all affine transformed to
MNI space. The version of MNI space is only really defined for affine
aligned brain images (ie the rather blurred averages). You should
note that the single subject brain that people often overlay their
results on does not define MNI space. It is just the scan of a single
individual that has been affine registered to MNI space.
On 24 February 2011 16:27, simon vandekar <[log in to unmask]> wrote:
> Thanks for the reply John,
> As far as I can tell the template*.nii files look good, they are gray matter
> images. There is also a u_rc1*_Template.nii in the same folder as my
> template*.nii files, I've attached the image for that. Is that what my flow
> fields are suppose to look like?
> Thank you,
> On Thu, Feb 24, 2011 at 10:47 AM, John Ashburner <[log in to unmask]>
>> It is a bit suspicious if all the u_*.nii are just zero. I'm not sure
>> why this would be the case. Do the template*.nii files seem to be OK?
>> Best regards,
>> On 24 February 2011 15:03, Simon Vandekar <[log in to unmask]> wrote:
>> > Hi John and SPMers,
>> > I am attempting to normalise my fMRI data using the steps described in
>> > pages 440 and 441 of the spm8 manual:
>> > 1. Slice timing
>> > 2. Realign: Estimate and Reslice
>> > 3. Coregister the structural to functional images
>> > 4. Registration looks good.
>> > 5. Segment the anatomicals the SPM5 routine
>> > 6. Initial Import- use the *seg_sn files
>> > 7. Run Dartel with dependency on the gray and white matter images output
>> > by import step
>> > 8. Normalise to MNI, select the template6 from dartel output, the
>> > subject's flow field, and normalise the ra* files (slice timing, resliced).
>> > The output I get is sdra* files. When I compare the output to the mni
>> > template it looks as if the subject's image has not been normalised. I am
>> > suspicious of the flow field because if I look at the individual subjects'
>> > uc1* image it is an empty black box.
>> > What am I doing incorrectly here?
>> > Thanks in advance,
>> > Simon