Hi Jack - ok, if you want to do this then the thing to do is to get a
func3->func1 transform via func3->struc3->standard->struc1->func1 (or miss
out the standard steps if you want to create your own transforms with
FLIRT outside of FEAT). Use convert_xfm to concatenate all these
transforms and then resample func3 to func1 space directly.
Cheers.
On Thu, 29 Jul 2004, Jack Grinband wrote:
> Hi Steve,
> I don't want to resample the funcs to standard space. I want to keep them in func space, but I
> want them to be aligned to each other.
>
> 1. The simplest solution is to register func2 and func3 to func1. Now you have all three funcs
> aligned and in the same space. The problem with this is that they are all low contrast to noise,
> low resolution images and you're almost guaranteed to have a sub-optimal registration.
>
> 2. The second solution is to register func1, func2, and func3 to standard and then apply
> standard2func1.mat to func2 and func3. Now they're all aligned to func1. The problem here is
> that we don't want to resample the funcs to standard space then back to func space.
>
> 3. The third solution is to register func1, func2, and func3 to standard but set: -ref func.
> Presumably, this should align all three without resampling. But this is where I ran into problems.
> After adding the extra offset to account for FOV, the brains were still not aligned.
>
> thanks,
>
> jack
>
> >HI - I hate to say it but you almost certainly _don't_ want to be
> >resampling the funcs "to" standard space but keeping them in func space,
> >for various reasons - what is it you're wanting to do this for?
> >
> >Cheers, Steve.
> >
> >> I have two func brains that were successfully registered to standard. I then applied
> >> func2standard without resampling.
> >>
> >> standard FOV=[182 218 182]
> >> func FOV=[200 200 93.5]
> >>
> >> diff = (func FOV - standard FOV)/2
> >> = [9 -9 -44.25]
> >>
> >> To adjust for FOV, I added diff to the x,y,z values of the two func2standard.mats. While
> >> they were mostly overlapping, they were off by up to 25 mm.
>
Stephen M. Smith DPhil
Associate Director, FMRIB and Analysis Research Coordinator
Oxford University Centre for Functional MRI of the Brain
John Radcliffe Hospital, Headington, Oxford OX3 9DU, UK
+44 (0) 1865 222726 (fax 222717)
[log in to unmask] http://www.fmrib.ox.ac.uk/~steve
|