Fix interaction between GDC and registration#143
Merged
Conversation
- Use eroded masks for histogram matching and registration to mitigate hypointensities at the peripheries of the masks leading to hyperintensities in the contrast-inverted images. - Perform contrast inversion using a linear rather than reciprocal transform, taking into account the prospect of values that are negative / zero / small in magnitude (which could be introduced by eg. gradient non-linearity distortion correction of the T1w) that have a deleterious effect on the reciprocal transform. Closes #142. Closes #141.
This was referenced Mar 4, 2026
|
Hi Rob, Tried running it last night and ran into the following Run command was: Running with a GPU, if that changes anything. |
Error in modifications in 54cff60.
Collaborator
Author
|
@aaroncapon Dumb mistake on my part; plz try again. |
|
Tested this last week and the results were checked by Phil. He said they look good. |
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Tagging @ppruc, @aaroncapon.
This looks to me to produce an adequate alignment; see below.
@ppruc Note that the black pepper artifact we saw in the pseudo-b=0 generated from the T1w is still present. This is due to the use of SynthStrip in deriving a DWI brain mask. Its priors on brain shape result in the inclusion of hypointense voxels at the inferior edge of the frontal lobes where there is signal dropout. When included in the intensity histograms to be matched, this pulls the brightest T1w voxels down to b=0 background intensity levels. This could likely be mitigated with some heuristic, but realistically I don't think this effect should be too deleterious to rigid-body registration.
I've only run these commands manually, so fingers crossed I've not borked the syntax somewhere.
Will be merged into #137 once confirmed working.