Dear Raphael,
I'm afraid these isn't much documentation for this module yet; it's on
the todo list though.
As for the adjustment option, see the help text in spm_regions.m:
> Responses are adjusted by removing variance that can be predicted by
> the null space of the F contrast specified (usually an F-contrast
> testing for all effects of interest).
In the batch interface, you have to enter the index of that contrast
from your list of contrasts (ie i from SPM.xCon(i)).
Best regards,
Guillaume.
On 24/11/14 15:10, Raphael F.C. wrote:
> Thank you, Guillaume! I didn't realize I could use more than one Region
> of Interest...
>
> And do you know if there is a documentation for this module? For
> instance, I am not sure of what I should use for the "Adjust data"
> option in the module.
>
> Regards,
> Raphael
>
>
> 2014-11-24 10:20 GMT-02:00 Guillaume Flandin <[log in to unmask]
> <mailto:[log in to unmask]>>:
>
> Dear Raphael,
>
> you should be able to do this with the 'Volume of interest' module.
> Create three regions of interest:
> * Thresholded SPM
> * Mask Image: your mask of hippocampus
> * Sphere
> Global maximum
> SPM index: 1
> Mask expression: i2
> and then finally use:
> Expression: i1 & i3
>
> Best regards,
> Guillaume.
>
>
> On 24/11/14 02:59, Raphael F.C. wrote:
> > Dear SPM experts,
> >
> > I am going to perform a DCM analysis in a group of images with SPM8.
> >
> > My goal: I have a SPM.mat file from "contrast manager" step and a mask
> > of hippocampus. This mask would be used to look for the voxel inside
> > hippocampus with the greatest t contrast value. Around this voxel I
> > would draw a sphere of a specific radius to extract the eigenvariate
> > time-series.
> >
> > I was searching how to do it in the internet and in SPM functions,
> but I
> > am still lost.
> >
> > What I have gathered so far:
> > 1) I can use: batch SPM->Util->Volume of interest, but I cannot add a
> > mask here to look for the voxel with the greatest t value and then
> draw
> > the sphere around it. Should I do this search outside the batch
> and then
> > fill the batch with the voxel address?
> > OR
> > 1) I should use spm_regions.m function to create a .mat file with the
> > eigenvariate time-series.
> > 2) The problem is that I have to use other functions like
> > spm_results_ui.m and spm_getSPM.m to create the necessary inputs for
> > spm_regions.m
> >
> > Any ideas?
> > All the best,
> > Raphael
> >
> >
> > --
> > Raphael F. Casseb
> > Medical Physicist, Ph.D. Student
> > Medical Physics Lab - State University of Campinas
> > Contact: +55 19 3521-8246 <tel:%2B55%2019%203521-8246>
>
> --
> Guillaume Flandin, PhD
> Wellcome Trust Centre for Neuroimaging
> University College London
> 12 Queen Square
> London WC1N 3BG
>
>
>
>
> --
> Raphael F. Casseb
> Medical Physicist, Ph.D. Student
> Medical Physics Lab - State University of Campinas
> Contact: +55 19 3521-8246
--
Guillaume Flandin, PhD
Wellcome Trust Centre for Neuroimaging
University College London
12 Queen Square
London WC1N 3BG
|