Hi - you could adapt Featquery to work on the lower-level results in
standard space - but it might be quite a lot of work to make these
changes.
However, I wouldn't think that this is necessarily a good idea. If
the mask is varying by such a large factor across subjects, then this
suggest that possibly the standard space registrations were not good?
Or alternatively is this because the masks are very tiny/thin and
that the downsampling and thresholding is having a large effect? In
which case you might want to change (eg lower) the post-interpolation
thresholding option.
Cheers, Steve.
On 27 Jul 2007, at 12:32, Daniel Irlam wrote:
> Hello
>
> I have a (hopefully simple) practical question regarding use of
> Featquery.
>
> At the moment I am creating masks in standard space defined from
> higher level (group) analysis.
> When I use such a mask in the Featquery it naturally gets
> transformed into the lower resolution
> subject specific space of each individual subject. Somewhat
> suprisingly (well to me at least)
> the size of the transformed subject-space mask can vary by a factor
> of 2 or more. Intuitively I
> don't like this extra degree of variability in my analysis.
>
> I wondered therefore whether it was possible in Featquery (or
> simple for me to change), that rather than
> transforming the standard space mask into individual subject masks,
> instead to invert the logic and
> use the subject space stats transformed into standard space with
> the mask? Having done a higher order
> analysis, the transformed subject space stats already seem to exist
> in my first level feat /reg_standard/stats
> dirs.
>
> many thanks
>
> Daniel
>
> --
> Daniel Irlam Ph.D
> [log in to unmask]
------------------------------------------------------------------------
---
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
------------------------------------------------------------------------
---
|