Thanks Saad. Just to see if I could get it to work, I tried running:
$ probtrackx2 -s BEDPOSTin.bedpostX/merged -m BEDPOSTin.bedpostX/nodif_brain_mask.nii.gz -x GM_WM_boundary.nii.gz -P 50 --omatrix1 --stop=GM_WM_boundary.nii.gz.nii.gz --forcefirststep --distthresh=3 --steplength=2
When I look at the fdt_matrix1.dot file, I get some behaviour I don't understand. Sorted on the seed voxel, the file starts with:
1 1 50
1 17 50
1 18 1
1 22 17
1 23 1
1 39 42
1 40 1
1 44 50
1 45 1
1 69 49
Now with --forcefirststep I understand I see the (1,1,50) tuple. But how is it possible that if all 50 streamlines end up in voxel 17 (which is both a seed and a stop condition), other voxels are also reached? If a streamline goes through, say, voxel 18 and then passes into 17, it should already have been terminated at 18. Why does the third column not add up to 50 (or 100 if we include the self-loop)?
Best,
Max
|