Since you are sending this to the FSL list, even though you are
addressing us as "SPM users", I will assume you are looking for a
suggestion based on FSL.
If you have a label image (i.e. an image volume that defines ROIs by
assigning nominal values to the voxels that belong to them) it is
straightforward to calculate the ROI volumes with fslstats. Example:
if you have three ROIs labelled 1, 2 and 3 on a background of zeros,
you can use the histogram feature:
fslstats labelimage -h 4
This lists the volume of each label in voxels, including background. Use
fslstats labelimage -h 4 | nl -v 0
to have the ROIs numbered in the output.
To get the volume of an individual ROI in mm^3, use the volume option
with thresholding. For ROI 2, use
fslstats labelimage -l 1 -u 3 -V
Without thresholding, -V will work for images defining a single ROI,
independent of the values of the voxels within the ROI. That means you
can use it for MR images where an ROI has been masked.
Hope that helps.
Rolf
On 11 July 2010 12:04, Alexander Lebedev <[log in to unmask]> wrote:
> Dear SPM users,
>
> Sorry for disturbing. Could you please advice me the best way (software)
> to create and calculate ROI volumes?
>
> My task is to calculate volumes of
> several CSF spaces.
>
> My tactics:
>
> 1. Segmentation in SPM (getting data for CSF);
> 2. Starting MRICroN and combining my original (T1) data with segmented
> image (in order to perform adequate navigation);
> 3. Easily creating a VOI and...
> getting stuck with it... because I have no idea what should I do
> next (as far as I understood, all tools in MRICroN are for group
> analysis)...
>
> Please, help me if it is possible...
> Thank you beforehand!
> --
> Sincerely Yours,
> Alexander Lebedev,
>
> I.P. Pavlov State Medical University,
> 197022, Lev Tolstoy str. 6/8,
> Saint Petersburg, Russian Federation.
>
|