bioformats
bioformats copied to clipboard
ome
DICOM: fix certain types of data with multiple optical paths
See https://forum.image.sc/t/dicom-folder-to-ometiff/61402/10. Test dataset is in inbox/imagesc-61402.
Without this PR, showinf -nopix -noflat on any of the files will show a pyramid with 5 resolutions and 12 Z sections. Examining the files with dcdump indicates that there should be 12 channels, not 12 Z sections.
showinf -nopix -noflat with this PR should correctly detect 12 channels and 1 Z section. Reading any of the resolutions with showinf -noflat -resolution r should successfully read all 12 channels, with obvious data in each channel.
Keeping this as a draft for the moment, since I haven't yet checked this against any of our other data.
This pull request has been mentioned on Image.sc Forum. There might be relevant details there:
https://forum.image.sc/t/dicom-folder-to-ometiff/61402/16
Repo test failures: https://merge-ci.openmicroscopy.org/jenkins/job/BIOFORMATS-test-folder/lastFailedBuild/consoleFull
Sorry, I didn't realize draft PRs would still be included in the builds. Adding the exclude label for now.
@melissalinkert if you need any help with this in terms of understanding the layout of the data, or validating any of your assumptions, please let us know - @dclunie is happy to help!
I believe this is now working correctly, so reincluding in the builds for verification.
The last issue was a subtle problem with how some edge tiles are handled, which should be fixed with 3881f0c. That requires a configuration change for the macro and thumbnail images in CPTAC-LSCC_C3L-00965-26_PixelMedConvertedSVSSuppliedPyramid, but showinf -noflat -crop 0,0,256,256 -series 1 with and without this PR will confirm that the previous behavior was incorrect.
Tests are passing with this change included: https://merge-ci.openmicroscopy.org/jenkins/job/BIOFORMATS-test-repo/1301/
@melissalinkert are those builds available for download so that the user who reported the problem in this thread could test and confirm this works for them?