Hello everyone, I had a general question regarding a specific use case.
So our use case right now that we want to integrate DICOM support into our application which uses OpenSeadragon as the viewer.
My thinking was that we use Orthanc as the DICOM based server and use it’s IIIF server capabilities hence we decide to move to better viewers in the future we can keep the backend the same.
The problem is when i try to view a DX modiality i.e a chest xray it’s view is sort of inverses.
My guess is that orthanc not applying the correct transformations before sending it to openseadragon.
I read in the tutorial that this plugin’s main use case is slide images but can be used with other modalities as well.
My question is that is this something that can be easily solved, i.e some hints would be super helpful like the correct functions to apply before sending the image. I am a c++ developer so I can get to it if provided some small hints.
Or is the problem something bigger or I am not understanding it right as well maybe.
Your help is greatly appreciated,
Thanks
The left image is viewed through openseadragon,
The right image is viewed through OHIF.
As a new user can only upload one image hence the cropped images
Hi,
Well, I’m actually not surprised by the image being so dark. All other viewers perform windowing adjustment after the image has been loaded and they let the user adjust windowing interactively.
I’ll let @jodogne give you more details about possible fixes. From my perspective, WSI plugin should only be used for WSI - but I’m not a specialist of this plugin.
Best regards,
Alain
1 Like
Thanks for the info, so the proper way is to preform these transformations at the viewer side I see.
I completely agree that it’s not the intended purpose but was just curious regarding our very unique use case. Thanks for the info, I will have a look if i can get that info from orthanc to preform at the viewer side.
@jodogne Hello, would you be able to provide some more insights, a very high level overview if possible.
Hello,
I confirm that the OrthancWSIDicomizer tool is designed for visible-light imaging, and makes the assumption of RGB images using 24bpp.
However, since release 3.0 of the WSI plugin for Orthanc, its Web viewer supports the on-the-fly generation of multi-resolution pyramids from any DICOM frame, without prior DICOM-ization to the “VL Microscopic Image” IOD. This corresponds to the “deep zoom” button that is present in the Orthanc Explorer at the instance level.
In releases 3.0 and 3.1 of the WSI plugin, this new feature didn’t take into account the windowing and photometric interpretation information that is present in the DICOM files. However, the next release 3.2 of the plugin will implement this rendering. This feature is now pending in the mainline.
If you provide me with a problematic sample DICOM image, I could give a try.
Regards,
Sébastien-
