How to get specific WM pathway (tracts) between two ROIs?

1

Good day to you all.

I am interested in the WM pathway (tracts) between two ROIs, e.g. AMG and ATL (spherical ROI).

tckgen wmfod.mif mytracks.tck -seed_image AMG.mif -include ATL.mif
then reversed:
tckgen wmfod.mif mytracks.tck -seed_image ATL.mif -include AMG.mif
Finally, the reverse-track pairs were combined to create the final tracts (how to do this step?)

Perform wholebrain tractography using the same tractography parameters as above (1).
then
tckedit -include ATL.mif -include AMG.mif

which one should be preferred? Or is the correct method not outlined above?

Thanks.

Thanks.

1 Like
2

You could use tck2connectome to get streamline indexes from a whole-brain connectome that connect your ROIs, and then connectome2tck to extract those streamlines as a .tck.

You just need to make a nodes.mif type of image. It is a single image where your two masks have different intensity values.

Best,
Steven

3

In your first option, if you just want to combine all tracks from both outputs, tckedit will do this for you.

tckgen wmfod.mif mytracks_forward.tck -seed_image AMG.mif -include ATL.mif
tckgen wmfod.mif mytracks_backward.tck -seed_image ATL.mif -include AMG.mif
tckedit mytracks_forward.tck mytracks_backward.tck mytracks_combined.tck

tckedit also takes ROIs if you want to narrow down what’s included in the final combined .tck file.

If you would like your tracks to begin and end in your two ROIs then you could try the tckgen options -seed_unidirectional and -stop, e.g.

tckgen wmfod.mif mytracks.tck -seed_image AMG.mif -seed_unidirectional -include ATL.mif -stop

will should give you streamlines that leave the AGM in one direction and terminate once they reach the ATL.
I’m not experienced with WBT so I couldn’t say which would be preferable in your case.

- Fiona

4

Hi @Fullcream,

It’s such a common operation, and it’s a common point of discussion on here, yet somehow we don’t really have documentation.

Rather than respond here I threw together a new Wiki page. Maybe people who actually do this kind of experiment on a regular basis can provide suggestions for imrovements?

Cheers
Rob

1 Like
5

@fionaEyoung @smeisler
Thank you for your answer! It’s helpful.
@rsmith
Thanks a lot, Robert. The new wiki page is very comprehensive :+1: :+1: and my problem has been solved.