potential impact on TCIA-related features

[kirbyju]

Hi, I just wanted to give you a quick heads up about some changes at TCIA which might be of interest for the integration features you've implemented. It appears you're pinned to a much older version of tcia_utils, and we don't have any immediate plans to phase out the old API (v1) that you're using in that release though it in the process of being deprecated.

In the process of dealing with that, we have also made some changes that should significantly speed up downloads if you upgrade tcia_utils. I just wanted to caution that, if you do this, the nbia.getSimpleSearchWithModalityAndBodyPartPaged() function you're using is now called nbia.getSimpleSearch(). I don't think the output that function produces has changed.

I don't know if this would matter for your purposes, but another significant change is that we no longer use NBIA for handling "controlled-access" datasets so all of the functionality surrounding the use of tokens and credentials has been removed from tcia_utils.

Finally, the base URL is also up to v4 now (I noticed a reference to v1 in your code).

Best,
Justin

[zjp]

Thanks for the heads-up! I have updated us to tcia_utils 3.2.1

[kirbyju]

Hey, nice to hear that! Just for longer term planning purposes, things have actually continued to evolve on our side since October and we will be deprecating support for the NBIA software that holds our DICOM data over the next year or two. We will be relying on the NCI Imaging Data Commons (IDC) to host our open-access public DICOM datasets and then anything that requires "controlled-access" will be handled by other parts of the NCI Cancer Research Data Commons. We'll be updating our API guides with more details about this as they're worked out.

For now, we've updated our Data Retriever software to make it so that regular users don't need to understand all this behind the scenes stuff if they're just coming to our site via a browser. For 3rd party integration apps like yours, it may make sense to work directly with the underlying platforms we'll be using.

https://github.com/ImagingDataCommons/idc-index should do most of what you're doing with the nbia.py module of tcia_utils. https://github.com/ImagingDataCommons/idc-claude-skill is also extremely helpful related to asking AI to help draft code to access IDC data. https://github.com/ImagingDataCommons/SlicerIDCBrowser is a very similar Slicer extension to what you're doing in ChimeraX in case that provides further inspiration.