Dear Yves,
Many thanks for your post! I am happy of reading you, because I think retrospectively that I have misunderstood Vinicius' initial question.
At that time, I was indeed focusing on 3D volume rendering (and not MPR volume rendering, which displays 2D slices). I was hoping that the emerging WebGL technology could enable full 3D volume rendering inside Web browsers... unfortunately, it has since appeared that people from the Khronos Group are not very keen on implementing primitives for native MIP rendering as an extension to WebGL 1.0:
Unfortunately, MIP rendering is necessary for clinical visualization (that should be as fast as possible). As a consequence, by contrast with my message from May 2013, my current conviction is that WebGL 1.0 is not mature enough for clinical visualization. WebGL 2.0 will support MIP rendering, but will only be supported by browsers in 1-2 years. So, I think that 3D volume rendering has currently to be implemented either as a standard, heavyweight, native client (such as OsiriX, Slicer or medInria that all use VTK), or as a native plugin for Web browsers (such as the NaCl technology from Google Chrome):
This also implies that full 3D volume rendering cannot currently be brought into Orthanc without endangering its lightweight philosophy. So, I am now convinced that Orthanc Explorer should embed a tool for lightweight MPR visualization inside Web browsers. And, of course, dwv is one very interesting candidate for that purpose!
To answer the second, technical part of your question, Orthanc does not currently implement WADO as such (it is on the mid-term roadmap), but implements a very similar mechanism as it provides a REST API to make the DICOM-to-PNG conversion on-the-fly. Concretely, provided the DICOM series of interest has the "XYZ" Orthanc identifier, you can retrieve the list of its images using a single GET request:
=> then, look in the "Instances" field to get the Orthanc ID of all the images of this series. This request can of course by done quite easily with AJAX.
To download a signed 16bpp PNG file of one of these images (whose Orthanc ID is, say, "ABC"), you would just use:
=> The result is the PNG image of interest. This is actually quite similar to WADO, but in a RESTful perspective.
The Orthanc REST API allows you to download the 8bpp PNG, the unsigned 16bpp PNG, or of course the raw DICOM file. The full reference of the REST API can be found at the following address:
So, I think that connecting dwv to Orthanc is clearly possible and would be a great idea. What do you think about this opportunity?
Do not hesitate to contact me if you have further questions about Orthanc.
Cheers,
Sébastien-