Hi - yes, I'm not that surprised that making these changes to the data
causes some threshold/mask problems.
After your changes, do you still have a background with values close to
zero? You probably need to. You may also need to rescale all values by
(eg) 1000 to solve various practical issues. If you want to make an
example FEAT directory available we can take a quick look and make more
concrete suggestions.
Thanks, Steve.
On Mon, 30 Jun 2003, Joshua Korn wrote:
> Hi all,
> For various reasons, I am converting my images into percentage
> changes before running FEAT, so "intensity" values are centered around 1
> and vary by a few percent. Obviously, I use floating points for this
> (data type 16). Unfortunately, feat appears to floor floats at some
> point in it's operation (as judged by the time series plot in the feat
> report), and I was wondering if anyone knew if this occurred after stats
> were computed but before plotting, or before stats were computed (in
> which case I need to change things around, I suppose...)
> Additionally, this gives issues with the mask feat craetes; is this
> mask created from the brain/background percentage that is set? How do
> you stop feat from creating any mask, or how do you provide your own?
> Thanks in advance,
> Josh
>
Stephen M. Smith MA DPhil CEng MIEE
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
|