Hi,
It seems most likely that the f-test problem really isn't related to
the different resampling of the first-level images. Are you _sure_
that the second-level model and contrasts (and f contrasts) are all
correct and valid? You might want to send them in.
In any case, an easy way to check that the resampling is ok is:
- make sure you have overwritten the example_func2standard.mat
transform in the .feat/reg directory with your Nudge output
- rm -rf reg_standard
- featregapply <featdirname>
Cheers, Steve.
On 25 May 2007, at 23:52, Paul Borghesani wrote:
> Hi -
>
> In order to correct minor problems with example_func2standard
> transformation I have used the
> nudge_gui to make a transformation matrix which I then applied to
> my reg_standard/stats folder.
> Thus I modified my reg_standard/stats copes and varcopes in my
> individual .feat folders. When I
> redid my higher level analysis with these modified files I get
> exactly what I expect; i.e. slightly
> different zstats in slightly different locations (frankly, quite
> minor changes after considerable
> work!). However, when I try to run an F test on these
> updated .feat folders I get the following
> error message (also see attached report.log):
>
> __________
>
> An exception has been thrown
> Runtime error:- detected by Newmat: matrix is singular
>
> MatrixType = Crout # Rows = 2; # Cols = 2
> Trace: Crout(lubksb); GeneralSolvI; InvertedMatrix::Evaluate;
> Gsmanager::f_ols_contrast;
> Gsmanager::ols_contrasts; Gsmanager::ols; Gsmanager::run.
>
>
> echo 4 > stats/dof
>
> /bin/rm -f stats/zem* stats/zols* stats/mask*
> Rendering using zmin=2.3 zmax=8
>
> _________
>
> Note: I am only using 6 .feat files inorder to de-bug this
> process. Running the same F test with
> the "pre-nudged" data works exactly as expected. Is there a reason
> why after modifying the
> reg_standard/stats folder that higher leval analyses in general
> work, but F tests do not? I suspect
> that it has to do with the other .feat files (copes, varcopes, dof,
> reg folder) that I have not updated
> with my nudge transform, but from reviewing the log files I can't
> figure out which of these might
> be a problem.
>
>
> Thanks for you help
>
> Paul Borghesani
> University of Washington
>
>
> <report.log>
------------------------------------------------------------------------
---
Stephen M. Smith, Professor of Biomedical Engineering
Associate Director, Oxford University FMRIB Centre
FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK
+44 (0) 1865 222726 (fax 222717)
[log in to unmask] http://www.fmrib.ox.ac.uk/~steve
------------------------------------------------------------------------
---
|