help > Correlating seed-to-voxel results to template
Showing 1-2 of 2 posts
Display:
Results per page:
Sep 23, 2019  01:09 PM | Andreas Voldstad - University of Oslo and LMU of Munich
Correlating seed-to-voxel results to template
Dear CONN users,

In the ICA results summary tools, you can correlate your ICA components to the template components provided by CONN, such as the Default Mode Network.

I am wondering if someone has any tips on how to perform similar correlations between seed-to-voxel T-maps and other spatial templates. Is there a function in CONN or some MATLAB code that could implement this?

Ideally, I would like to create a correlation matrix visualising the similarity between the T-maps from different seeds and templates.

Best,

Andreas
Sep 26, 2019  12:09 PM | Alfonso Nieto-Castanon - Boston University
RE: Correlating seed-to-voxel results to template
Dear Andreas,

Assuming that the different T-maps & templates all have the same reference space & voxel-sizes, you could do that using something like:

   myfiles = {'file1.nii','file2.nii','file3.nii'};
   vols = spm_vol(char(myfiles));
   data = reshape(spm_read_vols(vols),[],numel(vols));
   data(isnan(data)) = 0;
   R = corrcoef(data);

The resulting variable R contains the matrix (3x3 in this example) of spatial correlation coefficient values between each pair of files specified in "myfiles" 

Hope this helps
Alfonso

Originally posted by Andreas Voldstad:
Dear CONN users,

In the ICA results summary tools, you can correlate your ICA components to the template components provided by CONN, such as the Default Mode Network.

I am wondering if someone has any tips on how to perform similar correlations between seed-to-voxel T-maps and other spatial templates. Is there a function in CONN or some MATLAB code that could implement this?

Ideally, I would like to create a correlation matrix visualising the similarity between the T-maps from different seeds and templates.

Best,

Andreas