I think this probably makes sense, yes. Presumably the dof=3 option
doesn't work well because you don't have pure translation between your
FMRI data and the initial structural. It's quite possible for the other
options to give similar results to each other if your scanner is well
calibrated for the different image types.
See
http://www.fmrib.ox.ac.uk/fsl/feat5/detail.html#reg
for a little more detail.
Hope this helps? Cheers, Steve.
On Sat, 21 Aug 2004, Sarah Shuwairi wrote:
> hi steve!
>
> the html reports for each individual run were looking fine, however, when i combined multiple runs within a subject i had misalignment of activation maps with the template. i believe i've solved that problem for the meantime by doing 2 things first: (1) aligning all the functional and structural .hdr files, and (2) stripping all my files with BET before i begin the 1st level analysis.
>
> i still have an issue regarding the appropriate degrees of freedom for
> the 1st level analysis. the DOF options on the registration tab also
> seem to have affected my results: using 3 - 6 - 12 yields misalignment
> of activation; but 6 - 7 - 12, and 7 - 12 - 12 both yield pretty much
> the same pattern and magnitude of activity in analogous brain locations,
> which is good! but, i'll be doing another analysis across subjects, so
> is one of these choices better or more accurate to use in the 1st level?
>
> thank you,
>
> sarah
>
>
> ----- Original Message -----
> From: Stephen Smith <[log in to unmask]
> ox.ac.uk>
> Date: Wednesday, August 18, 2004 5:47 pm
> Subject: Re: Registration Problems in FSL
>
> >
> > The thing to check here is the registration report within the FEAT web
> > page report. Do all of the stages of the registration look ok?
> >
> > Cheers, Steve.
> >
> >
> > On Mon, 16 Aug 2004, Sarah Shuwairi wrote:
> >
> > > Hello Steve,
> > >
> > > I am writing about some registration difficulties I am encountering
> > > with a higher level analysis aimed at combining 3 runs from 1
> > subject> in a single scan session. Using the GUI, I've done a
> > separate lower
> > > level feat analysis for each of the 3 functional runs, and each
> > one has
> > > been registered to the subject's T1 inplane structural image(
> > with 3
> > > dof, translation only), and then to the 3D high res structural scan
> > > (with 6 dof, full search), as well as to the MNI template (12
> > dof, full
> > > search). However, after the runs are combined and viewed in MNI
> > space> the significant activations are nearly all appearing
> above
> > the brain
> > > tissue space (this occurs in both the html report and the threshold
> > > images for each of the contrasts). My images were initially
> > acquired> as 32 whole-brain axial slices, 3x3x3 mm voxels.
> > >
> > > Is there something I ought to be doing differently at a particular
> > > stage of analysis to adjust the images to another format or size
> > to get
> > > things correctly oriented and registered in the end?
> > >
> > > Thank you for your help,
> > >
> > > Sarah Shuwairi
> > > Graduate Student, NYU
> > >
> > >
> >
> > 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
> >
> >
>
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
|