Update @itk-wasm/dicom for new SEG pipeline

PaulHax commented 1 week ago

[x] Deal with overlapping segments coming out as multi-component images. Short putt: turn mergeSegments option to true, pull first output file/component for use as segment group, and pop a warning.
[x] Stash segment label and color tags in a store on load. Then create new segments with colors when converting SEG volume to segment group.

closes #639

Suggest we just put these in an issue and not do in this PR:

[ ] Fix thumbnail for big SEG here: https://github.com/Kitware/VolView/issues/639
[ ] Loading indicator. Like a spinner on the image in the image browser.
[ ] When loading SEG volume without "background" volume, apply approate color map, surface segment group colors and labels to GUI.
[ ] Handle overlaping segments.

netlify[bot] commented 1 week ago

Deploy Preview for volview-dev ready!

Built without sensitive environment variables

Name	Link
Latest commit	445fa47fa06182905c8089cf68a907c5201e86db
Latest deploy log	https://app.netlify.com/sites/volview-dev/deploys/66ff21e7fc3c1c0008ce0ca2
Deploy Preview	https://deploy-preview-653--volview-dev.netlify.app
Preview on mobile	Toggle QR Code... Use your smartphone camera to open QR code link.

To edit notification comments on pull requests, go to your Netlify site configuration.

fedorov commented 2 days ago

Short putt: turn mergeSegments option to true and expect one component image as output.

Just to make sure there is no misunderstanding here. mergeSegments option will merge non-overlapping segments into minimal number of files. So if SEG does not have any overlapping segments, unless there is a bug, you should get only one file when mergeSegments is on. But if some segments overlap, they will be saved into separate files with mergeSegments on. TotalSegmentator should not produce overlapping segments. But you can look at the NSCLC-Radiomics or QIN-Prostate-Repeatability collections that contains overlapping segments:

@jadh4v - one of the above would be a good sample to include in testing in https://github.com/InsightSoftwareConsortium/ITK-Wasm/issues/1208.

As always, to download the images in the above example, you can use StudyInstanceUID available in the URL and idc-index package:

$ pip install idc-index --upgrade
$ idc download 1.3.6.1.4.1.14519.5.2.1.3671.4754.121472087445374646718121301133

Note that to see overlapping segments in the above I used OHIF v2. The current OHIF v3 cannot display overlapping segments as of yet, see https://github.com/OHIF/Viewers/issues/3496.

PaulHax commented 2 days ago

I did misunderstood. Thanks for the clarification and files.

Changed to do the wrong thing better. That is: just using the first output file and popping a warning.

Kitware / VolView