You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
With MRtrix3 there's been a few requests over the years to have, for streamlines propagated bidirectionally from the seed point, knowledge of which vertex in the output streamline corresponds to the seed location. See e.g. tckmap -output_seeds option (which was at the time the easiest hack to satisfy demands).
In the context of this format, it's a DPS of either a uintX index or a 3.floatXX, depending on whether the seed point itself or an index in the output streamline is provided (theoretically it's possible that in some implementation the precise seed location won't actually be included in the output streamline; in MRtrix3 we guarantee that it's included, but I'm thinking of the most general case).
Primarily the question is whether or not such information should have some form of reserved keyword so that all implementations know where to write it in order to be trivially accessible / interpretable at read time.
The text was updated successfully, but these errors were encountered:
I agree we could have a list of reserved keywords, but it should also come with a small description and specification.
Seeds are definitively on my list too. It would be interesting to see what comes up.
With MRtrix3 there's been a few requests over the years to have, for streamlines propagated bidirectionally from the seed point, knowledge of which vertex in the output streamline corresponds to the seed location. See e.g.
tckmap -output_seeds
option (which was at the time the easiest hack to satisfy demands).In the context of this format, it's a DPS of either a
uintX
index or a3.floatXX
, depending on whether the seed point itself or an index in the output streamline is provided (theoretically it's possible that in some implementation the precise seed location won't actually be included in the output streamline; in MRtrix3 we guarantee that it's included, but I'm thinking of the most general case).Primarily the question is whether or not such information should have some form of reserved keyword so that all implementations know where to write it in order to be trivially accessible / interpretable at read time.
The text was updated successfully, but these errors were encountered: