Dear gPPI experts,

I'm now creating the VOI for the gPPI analysis. I want to create the VOI using one cluster mask or the independent atlas. But the create_sphere_image script only allows the construction of the spherical mask. I'm wondering is there any possible ways to create the 'image' VOI. Many thanks for your help.

All the best,
Tina

hi, Tina
I remember Marsbar can make a cluster mask from the SPM results. You can have a try.

Kitty

Dear Tina,

Kitty is right - if you want to use a cluster ROI, MarsBaR is a nice option: http://marsbar.sourceforge.net/
In MarsBaR, you would choose ROI definition -> Get SPM cluster(s)... Then choose your analysis SPM.mat file, threshold appropriately, and finally click on Write ROI(s) -> Write one (or all) clusters.
The ROI will be in .mat format, so you will need to transform it into a .nii file. To do this, you can do ROI definition -> Export.. -> Export ROI(s) to... -> image. Then pick the .mat file for your cluster. As space, I would choose 'From image' and choose an image that is in the same space as your gPPI analysis. In this way, you ensure that the ROI will be in the right space.

To use an independent atlas ROI, you can simply use any .nii file where all voxels in the ROI are 1, and all others are 0. For example, the Anatomy toolbox allows you to export such anatomical ROIs easily: https://www.fz-juelich.de/inm/inm-1/DE/F...

I hope this helps!

All the best,
Phil
----------------------
Philipp Kuhnke, M.Sc.
Ph.D. Student
Max Planck Institute for Human Cognitive and Brain Sciences,
Leipzig, Germany

Hi Kitty,

Many thanks for the answer. I just realized that just keeping the VOI dimension consistent with my data's dimensions, and then the gPPI analysis would run well. So after this question, I had made the VOI using the specified atlas in Anatomy toolbox and resliced it based on my imaging data's dimensions. The gPPI analysis get well results. Many thanks.

Hi Phil,
     Very clear and easy to follow. Many thanks for your nice suggestions.
     After this question, I had made the VOI using the specified atlas in Anatomy toolbox and resliced it based on my imaging data's dimensions. The gPPI analysis get well results.

You're very welcome! I'm glad you're getting some nice results too :)