Is it normal that there is almost no connectivity between the Vermis_CrusI (Region of Interest 9) and other Regions of Interest (ROIs) in the connectivity matrix generated when performing cerebellar structural connectivity analysis using MRtrix3 with the SUIT parcellation template?
Assuming that this is not a direct result of registration problems, I think I’ve had situations in the past where I’ve tried to incorporate sub-divisions of the cerebellum only to have some of them disconnected. It depends on the quality of the image data (the fossa can be finer than the imaging resolution), the quality of the tissue segmentation (notoriously hard for cerebellum), and how much DWI signal anisotropy is present there and suitably attributed to the WM-like ODF. It’s quite likely that just running tools off the shelf on conventional image data and expecting robust intra-cerebellar structural connectivity is going to yield disappointment. It will require some careful consideration of the limitations of tissue segmentation and tractography in that region.
Regards
Rob
Thank you for your reply.