I have some doubts regarding the cutoff value for the function tckgen. In the documentation it is written that “set the FA or FOD amplitude cutoff for terminating tracks (default is 0.1)”.
If I use the CSD model with probabilistic or deterministic tracking, is the cutoff value always referred to the FOD amplitude threshold? And is it referred to the FA threshold only if I don’t use CSD?
Is the cutoff value used to terminate the tracking when the FOD amplitude is below the value? Or is it used to remove directions from the probability mass function for selecting the tracking direction (in the case of probabilistic tracking)?
Yes, that’s correct - aside possibly from the FACT algorithm, where the cutoff presumably refers to the length of the vector (@rsmith will no doubt correct me if I’m wrong).
With deterministic tracking, it is used for termination. For probabilistic tracking, it serves both functions you suggest, but the latter is the more correct way of phrasing it: directions whose FOD amplitude is lower than the cutoff are not considered - they are effectively removed from the PDF being sampled. When no direction above the cutoff can be found within the allowed solid angle, then the streamline is terminated.
I am new to Mrtrix and trying to do some tractography on multi-shell data.
With respect to cutoff values in tckgen, the mrtrix webpage shows
-cutoff value set the FOD amplitude / fixel size / tensor FA cutoff for terminating tracks (defaults are: 0.05 for FOD-based algorithms; 0.05 for fixel-based algorithms; 0.1 for tensor-based algorithms).
However, the terminal command tckgen shows:
-cutoff value set the FA or FOD amplitude cutoff for terminating tracks (default is
0.1).
My question is, what is the actual default value of -cutoff in tckgen when using iFOD2.
Also, could you please comment on how to determine the optimal value.
With respect to cutoff values in tckgen, the mrtrix webpage shows …
However, the terminal command tckgen shows: …
This suggests that your MRtrix3 installation is not up-to-date. The online documentation is generated from the command itself, just as is the terminal command help page; but the “standard” documentation linked to online is that generated from the most up-to-date code. You are specifically missing this change, which means that your installation is prior to version 3.0_RC3, and therefore includes a bug in the tracking code that was fixed in the release of 3.0_RC3, so we would definitely recommend updating your installation.
Also, could you please comment on how to determine the optimal value.
If we could determine the optimal value, that’s what we’d use
You can see in my montage here the different factors that may influence what cutoff value is deemed appropriate, and how changing the cutoff changes the appearance of the tractogram. It’s hard to concisely summarise the process of determining whether the cutoff needs to be increased or decreased. It’s something like: are the streamlines trajectories too liberal (both jumping between crossing fibre populations, and spatially extending further than they should), or the converse?
