I know that for med dream there is actually a custom Python script that returns some customized data used by the MedDream viewer. Doing something similar for OHIF might be an option, but I’d have to dig into the details. Just optimizing the DicomWeb config might be sufficient.
You also might want to look at this section regarding support for OHIF since that forum might be another place to ask your question:
If you really need to spit a “dicom-json” file that is suitable for OHIF, as indicated by Stephen, creating a Python plugin for Orthanc is clearly the way to go: https://book.orthanc-server.com/plugins/python.html
Another point, if you want to use the Docker setups provided by Osimis as said by Stephen, but with ohif v3, the only working branch right now is the fix/v3-stable-docker.
Unfortunately these solutions force me to expose DICOMweb which is what I’m trying to avoid. The DICOM JSON integration allows me to tokenize and secure access to individual studies. I ended up writing out an endpoint to process a study into JSON format with a one-time access token to grant access to the JSON study data. I appreciate everyone’s input here though.