Dear Mrtrix3 community,
I would like to ask a question about group template analysis. I used the population_template function and FA, b0, and segmentation maps from all subjecctcs to successfully formed a group template.
Is there any way that mrtrix3 could help further generate group-averaged DWI, tensor or FOD, and eventually generate fiber tracking based on this group-averaged DWI?
I tentatively used mrcat to concatenate all DWI volumes that are deformed to group-template to make the group-averaged DWI. I then ran fiber tracking on top of that concatenated DWI images. I wonder if you would think this approach is reasonable.
Many thanks and very best,
Yixin
Hi Yixin,
One option might be to use population_template to generate a group-average wm-FOD (similar to what is done in the FBA pipeline, see here for single-shell and here for multi-shell data).
To get the FA maps into template space, store the warps generated from population_template and apply them after generating the population template (see -warp_dir option of population_template)
Hope this helps!
Best,
Arkiev
1 Like
Dear Arkiev,
Thank you so much for the suggestion! Rather than building a group-averaged wm-FOD. Do you think it would be possible to build a group-average Tensor?
Many thanks and very best,
Yixin
Hi Yixin,
While I haven’t tried this myself, I think it is possible to also use population_template to generate a group-average tensor image (using tensor images in subject space as inputs).
It might also be possible to use the FA maps to generate the group averages (see here for example usage).
Hope this helps!
Best,
Arkiev