Hi Brian,

If I am understanding correctly you would not want to normalize(write) those images (as they are already in MNI/normalized space, the SBC results image simply has been cropped to a smaller bounding box, one still encompassing your analysis mask, just to save space). This assumes that the data used in your SBC analyses had gone through preprocessing and as part of that it had been normalized to MNI space. If that is the case then you may simply want to load the data using a common reference (voxel size and bounding box) in order to compute the correlations, e.g. using something like:

file1 = 'template_networks.nii';

file2 = 'SBC_results.nii';

data1 = conn_vol_read(file1); % 91x109x91 data in file1

data2 = conn_vol_read(file2, file1); % 91x109x91 data in file2

mask = ~isnan(data1)&~isnan(data2); % disregard nan's

r = corrcoef(data1(mask), data2(mask)); % computes spatial correlation

Hope this helps

Alfonso

Originally posted by Brian Kavanaugh:

Hello,

In trying to examine the correlation between SBC maps and templated networks, I realized the networks.nii is 91 109 91 and the SBC results are 72 92 78. 

Would anyone know how to use spm's normalize(write) function to allign the dimensions?

Thank you!

 

brian