Deterministic or probabilistic tractography in ExploreDTI?

[Joost]

Dear Alexander and/or other ExploreDTI experts,

I am little confused about the function WholeBrainTrackingCSD_fast_exe_CL in ExploreDTI 4.8.6. 

As far as I understand, "DTI vs. CSD" and "deterministic vs. probabilistic" are two different choices, e.g. CSD does not necessarily imply that one is also using a probabilistic approach and vice versa. Thus, one can perform deterministic tractography using CSD, right? And from what I understand from the manual as well as the ExploreDTI forum, the probabilistic approach is currently NOT implemented in ExploreDTI (based on Alexander Leemans' reply in https://groups.google.com/d/msg/e_dti/lp6QohHsGgs/x384dTtRvNYJ and the fact that the probabilistic approach is not mentioned in the change logs in versions released since that date).

So, just to be sure: when I run WholeBrainTrackingCSD_fast_exe_CL, I actually perform deterministic tractography using CSD, correct? 

And for my full understanding, to which of the different tracking algorithms in MRtrix is this comparable (see http://mrtrix.readthedocs.io/en/latest/reference/commands/tckgen.html)?

Looking foward to reading your reply!

Best regards,

Joost de Jong

[Joost]

Anyone?!

Op maandag 5 februari 2018 15:32:18 UTC+1 schreef Joost:

[Alberto De Luca]

Dear Joost,
you are correct, the tractography in ExploreDTI is deterministic, so it is not directly comparable to the MRtrix implementation. By default the CSD approach uses recursive calibration, which is similar to dwi2fod of mrtrix with the "tax" option.

Best Regards
Alberto