Hi,
On 27 Sep 2007, at 22:28, Jonathan Hakun wrote:
> I'm using featquery to obtain the mean percent signal change (m-%
> sc) within
> a region (say 100 voxels large defined by a mask) for a single subject
> during 3 different tasks.
>
> I then want to draw a simple bar-graph showing m-%sc in each region
> separately during each task, not averaging across tasks.
>
> I'm assuming I need to show not only m-%sc within the region in my
> graph,
> but also show error bars for the standard error of the m-%sc within
> the ROI
> (this being a single subject S.E. would be relative to variation
> voxel to
> voxel, within the ROI converted to Percent-S.E. of m-%sc).
>
> Here comes the question...how do I calculate S.E. of the m-%sc
> within an ROI
> from Featquery output?
>
> I have some ideas, maybe you can steer me from there:
>
> I figure I can derive the Std Devation (S.E. to follow) of the mean
> Parameter Estimates from the values listed in the file
> "mean_mask_ts.txt"
> ...of course that's what I think I'm looking at... There are an
> equal # of
> values in that list to the # of volumes acquired in my task, and that
> "mean_mask_ts" has values roughly half way between the
> "max_cope1_ts" and
> "max_cope2_ts," so I'm just assuming it represents the mean of
> Parameter
> Estimates of all conditions for that 100 voxel ROI.
I doubt this is what you want, as I presume you want the STD across
voxels not timepoints. You could get a measure of the variability
across the voxels by looking at the 10-90% range in the Featquery
output table - or you could amend the featquery script to show you
the STD across voxels in a new column.
Cheers, Steve.
> I need help there...
>
> And further:
>
> If I'm in the right track, then once Parameter Estimates (automated in
> Featquery) have been converted to %sc, I'm thinking I need to
> convert this
> newly created S.E. of Parameter Estimates to "%" S.E. of Parameter
> Estimates. Is this as obvious as it sounds?
>
> I appreciate your collective time,
> Jonathan
------------------------------------------------------------------------
---
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
------------------------------------------------------------------------
---
|