New atlas from unfoldreg subfields #487
Open
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
I removed this in a clean-up PR, but it is actually needed.. Moved it to another location now..
- this adds a new form of atlas creation where the subject subfields come from "unfoldreg", i.e. from some existing base atlas (like multihist7), instead of from "native" (manual segmentations). - a --new-atlas-subfields-from CLI option has now been added to specify either native or unfoldreg - this supports the use-case of building a new atlas using additional metrics and on populations, the idea being that such an atlas should improve inter-subject alignment - atlas CLI args have also been renamed/simplified for clarity, adding --new-atlas in front of args that relate to new atlas building.. - dry-run test cases have also been included - note: behavior for existing atlas creation from native subfields should be unchanged (but you will now need the --new-atlas-from native)
akhanf
force-pushed
the
new-atlas-unfoldreg
branch
from
fb20959 to
2b7537e
Compare
now only disabled when modality==dsegtissue and group_create_atlas
Member
Author
|
Atlas generation is working now, and a newly made hcpUR100myelin template has been generated and pushed to hippunfold-atlases (uses curvature, gyrification, thickness, myelin).. note this atlas must be used with I've also confirmed that using a different atlas results in different metric maps, due to differences in the unfolded registration. Will demonstrate this in more detail for the paper.. |
native atlas gen still combines left and right together..
This reverts commit e32db2a.
was having dag issues so this simplifies the change to only affect the 2d unfolded registration
Collaborator
|
I'm happy to merge this in, but I was wondering if you're still planning to make a demo for the paper? |
Member
Author
|
Sorry I've been swamped with things haven't been able to bring my attention back to this.. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
I've been meaning to incorporate this feature for some time now, and the changes ended up being pretty minimal (ie one new rule) since most of the existing infrastructure (e.g. generic unfolded registration with any metrics, atlas generation with custom metrics) already supports it..
Still a WIP draft PR, as it is only dry-run tested, but I plan to build a HCP UR100 template including myelin as an initial use-case..
"unfoldreg", i.e. from some existing base atlas (like multihist7),
instead of from "native" (manual segmentations).
either native or unfoldreg
metrics and on populations, the idea being that such an atlas should
improve inter-subject alignment
in front of args that relate to new atlas building..
should be unchanged (but you will now need the --new-atlas-from native)
also includes a bugfix from a file I removed that was still needed..merged already