Repository navigation
Stain extraction: use a less strict condition across channels when thresholding - #316
Merged
Merged
Conversation
I think we want to allow pixels with substantial attenuation in only a subset of channels.
jakirkham
approved these changes
Jul 28, 2022
Member
|
@gpucibot merge |
Member
|
Thanks Greg for the PR and Gigon for the review! 🙏 |
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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 think we want to allow pixels with substantial attenuation in any of the channels rather than requiring substantial attenuation in all channels. The Macenko et. al. reference does not actually explain how the threshold was applied (min, max or mean across channels). However, qualitatively, results looked closer to expected with the
anycondition proposed here when I was working on a pending stain normalization example/demo.I have also seen third party implementations where there is an RGB->LUV conversion and thresholding is done on the L channel rather than in luminance space. That seems reasonable and perhaps is more intuitive in choice of the threshold number (easier to think in fraction of brightness rather than absorbance units). However, making that change would be a bigger deviation from the publication and would involve some small additional overhead of the RGB->luminance conversion.
I have marked this as breaking as it will affect the estimated stain vectors produced by
stain_extraction_pca.