Dear Paul,
I'm afraid that you cannot run the individual corrected outputs through first_utils
with the --saveVertices option. This is because first_utils requires a bvars input
when using this option. However, the vtk files are simple text files and if you look
at the *_Fthresh.vtk file then you'll see a list of numbers underneath the
lines that include "SCALARS Scalars float". These are the thresholded F-values
and they should be in the same order as the vertices output by the --saveVertices
option. So you can take these values and use them as a mask (note that the
values above 0 are not equal to 1 but equal to the F-values). This should allow
you to do the kind of plots you are talking about.
All the best,
Mark
On 19 Jun 2012, at 02:40, Paul wrote:
> Hi Mark
>
> Basically, i also want to plot some 2D correlation figures from this correlation analysis ( Instead of whole vertices correlation, i want to only focus on investigating the relationship between the displacement distance of statistical significant differences vertices and clinical evaluation), so i want to make sure and realize the step of the vertex based correlation analysis. Are there any commands or scripts in FSL can threshold the group comparison results (like using cluster tool to threshold the 1-p map under corrected P <0.05)? Follow threshold the statistical results from group comparison, could i simply use --savevertices flag with thresholded VTK file to get the coordinate information from these statistical significant vertices ? Finally could i calculate the displacement distance of each vertices (From each vertex coordinates of participants to the normal control mean coordinates or model coordinates which provided in FSL) ?
>
> Thanks !
>
> Paul
>
>
|