Hi all,
I’m still a bit new to tractography, so I apologize if my uncertainty is a bit naive. I’m working on a project comparing the performance of different diffusion MRI signal models (specifically DTI and free water corrected DTI) at tracking white matter pathways through edema with FACT. The version we’ve been using is version 3.0.4.
For tractography, the commands we used were as follows:
DTI FACT:
tckgen -algorithm FACT primary_vector.mif output.tck -seed_image FA_map.mif -select 10000 -cutoff 0.5 -seed_cutoff 0.6
DTI-FW FACT:
tckgen -algorithm FACT primary_vector.mif output.tck -seed_image fwcorr_FA_map.mif -select 10000 -cutoff 0.5 -seed_cutoff 0.6
For clarification, the “primary_vector.mif” file is exactly the same in both cases, and the only difference is the image file used in the -seed_image option.
Our understanding of what -seed_image does is that it allows users to input a mask image from which streamlines can be seeded entirely at random. Yet, knowing this, we observed a noticeably different result, where in the free-water corrected case we observed far more streamlines within the edema than in the standard DTI case.
How would this be possible if -seed_image is supposed to seed streamlines entirely at random? Is there any function within the workings of the -seed_image option that would cause such a drastic change in streamlines generated?
Any help or insight on this would be greatly appreciated.
Many thanks.