WIP: feat 194 class based color settings - bulk annotations

[pedrokohler]

Adding the ability to edit annotation groups colors based on their categories and types. Solves #194.

Screenshare.-.2024-06-10.12_43_53.PM.mp4

[github-actions]

Visit the preview URL for this PR (updated for commit 559ebce):

https://idc-external-006--idc-slim-preview-kk0jkjse.web.app

_{(expires Fri, 02 Aug 2024 13:27:32 GMT)}

_{🔥 via Firebase Hosting GitHub Action 🌎}

_{Sign: 88aacecd98ba54d2f9c8d201a9444e43d1ad8307}

[fedorov]

I am unable to point the preview instance to a GCP DICOM store - I get this error:

[pedrokohler]

I am unable to point the preview instance to a GCP DICOM store - I get this error:

@fedorov can you give me an example of a server url you're trying to use?

[fedorov]

projects/idc-dev-etl/locations/us-central1/datasets/idc/dicomStores/annotated_rms-original-and-annotations-wc-temp-20240122

[pedrokohler]

projects/idc-dev-etl/locations/us-central1/datasets/idc/dicomStores/annotated_rms-original-and-annotations-wc-temp-20240122

You have to add a /dicomWeb at the end (if you haven't)

When you do that we're getting a 401 error, probably because this dataset requires google authentication but the config file the environment is using does not ask you to login with google.

We're seeing how to change this right now... I'll let you know.

[fedorov]

Yes, you also should add the prefix. The full url is this: https://healthcare.googleapis.com/v1/projects/idc-dev-etl/locations/us-central1/datasets/idc/dicomStores/annotated_rms-original-and-annotations-wc-temp-20240122/dicomWeb. I just copied from the console. What I tested and what didn't work for me had both the prefix and suffix.

[igoroctaviano]

@pedrokohler whats the state on this one?

[pedrokohler]

@igoroctaviano as mentioned in slack I was not able to use the "live" channel id... however we can still try some other custom name. I had used idc-slim-preview for the channel id and it seems to generate a fixed URL, even though it has random bunch of characters at the end, I think it does not change. At least it hasn't with my tests.

We can give it a try and add the URL to the GCP permissions so we can try to authenticate.

[igoroctaviano]

@igoroctaviano as mentioned in slack I was not able to use the "live" channel id... however we can still try some other custom name. I had used idc-slim-preview for the channel id and it seems to generate a fixed URL, even though it has random bunch of characters at the end, I think it does not change. At least it hasn't with my tests.

We can give it a try and add the URL to the GCP permissions so we can try to authenticate.

Sounds good, please address the tests so we can merge.

[pedrokohler]

@fedorov this can now be tested with the dataset you required. Please do and let us know if we can merge.

[fedorov]

Per discussion in slack, it appears that currently this feature supports coloring of the annotation groups that come from bulk annotation modality (ANN). It should be able to handle annotations that come from SR, and allow per type/category coloring.

This is a representative sample that can be used to drive the development: https://idc-external-006--idc-slim-preview-kk0jkjse.web.app/studies/2.25.155017484756498730492136597238994838876/series/1.3.6.1.4.1.5962.99.1.3434988325.342625608.1687062201125.4.0.

[fedorov]

Per discussion in slack, I suggest we do the following. Let's shelve this current PR for now, and start a new PR that adds these features for SR annotations, into a dedicated new section, reusing as much from the current PR as makes sense. At a later time, when there is a need, we can come back to the ANN PR, and either add its features into a new section, or merge with the SR annotations section.

As @pedrokohler said, it's a good compromise... it won't make us waste what we have and we can make progress without much difficulty.

[sonarcloud]

Quality Gate failed

Failed conditions
32.8% Duplication on New Code (required ≤ 3%)

See analysis details on SonarQube Cloud

[igoroctaviano]

@pedrokohler whats the status of this pr? can we close it?

[pedrokohler]

@pedrokohler whats the status of this pr? can we close it?

Same status as the one explained in Andrey's last comment. I don't think it's supposed to be closed.

[igoroctaviano]

@pedrokohler if this is going to be open/maintained I suggest updating it from time to time and addressing conflicts as soon as they happen.