Dear FSL Experts,
I have have two groups of subjects I wish to compare with respect to "connectivity" between one subcortical and one cortical region using probtrackx2. The hypothesis is that one group would have a lower connectivity value than the another. Readings this forum has been very helpful, but I want to make sure I'm doing this right.
Cortical and subcortical regions were derived from individual T1s via FreeSurfer and were registered to each subject's native DTI space. I ran probtrackx with my subcortical ROI as the seed and the cortical ROI as the target. The values in the seed_to_target nii file seem to be what I'm after for my analysis. I could then use fslmaths to get the average of the streamlines going from voxels within the seed that reach the target. This is the valued I plan to compare between groups. I wonder though what I should do in terms of normalizing this value, or if that is even necessary for this type of analysis. I've read about using the waytotal for this (seed_to_target / waytotal). I understand using that approach for normalizing the fdt_paths file, but would this also apply to the seed_to_target values? Seed_to_target values within each voxel appear to be have a ceiling of the total number of streamlines sent (e.g., 5000 in default setting) which are quite a bit lower that the strongest values found in fdt_paths.
So, is it necessary to normalize the seed_to_target results in some way for group comparisons? If so, how is this best achieved?
Thank you for your help with this. I greatly enjoy working with FSL,
P
|