CSD on single-shell diffusion data: should this be done with CSD or SS3T instead of MSMT-CSD?

[MRTanja]

• What is the current behaviour?

1. In 02_proc-dwi.sh, constrained spherical deconvolution seems to be done with the same method regardless the number of shells available in the data. However it seems that MSMT-CSD is only appropriate for multi-shell data, not for single-shell. For example here https://community.mrtrix.org/t/single-shell-vs-single-tissue/5761/6 it says "If you have good multi-shell data, then you can use MSMT-CSD [...]. If you have good single-shell (+b=0) data, you can use SS3T-CSD". A distinction should be made in the pipeline between single-shell and multi-shell input. In the documentation it does say "This processing pipeline has been optimized for multi-shell DWI, but can also handle single-shell data", however I can't see where in the code in 02_proc-dwi.sh a distinction is being made.

2. On a related note, in the documentation for diffusion MRI it says "For multi-shell data, the Dhollander algorithm is applied to estimate response functions of CSF, gray, and white matter, and multi-shell, multi-tissue constrained spherical deconvolution is used to estimate fibre orientation distributions. For single-shell data, the Tournier algorithm and single-shell, single tissue constrained spherical deconvolution are used for these processing steps, respectively." but looking at the code it seems like only the Dhollander algorithm is used. I'm assuming this is fine even for single-shell data as was explained e.g. here: https://community.mrtrix.org/t/single-shell-vs-single-tissue/5761/6 but in that case the documentation does not match the code here.

• What is the motivation / use case for changing the behaviour?
  The method chosen to apply CSD to single-shell diffusion MRI data does not seem to be appropriate. As far as I know, this is only the right method for multi-shell data, while for single-shell data csd or ss3t_csd should be used.