Is the project still alive?

[Michel Casabianca]

Hello,

We, at DESKi, have been using Cornerstone to view echocardiographic exams for few years now. But last release was done on 2018-10-25 and there were only 10 commits since then. Thus I was wondering if the project is still maintained.

If this is not maintained anymore,  isn't that a security issue?

Next, I was wondering if there are Open Source or commercial alternatives.

Thanks

[Erik Ziegler]

Hi Michel,

Yes Cornerstone and related projects and libraries are still alive and well! There are hundreds of projects in academia and industry around the world that are currently building applications on our framework. Please keep in mind that we are a small team and have very limited resources and a lot of projects to maintain. If you are looking solely at commits to one repository (Cornerstone Core) you only have a small view into our overall framework. We also created and support the Open Health Imaging Foundation (OHIF) Viewer (https://github.com/OHIF/Viewers/) and its subcomponents (e.g. https://github.com/OHIF/react-vtkjs-viewport), as well as various other aspects of the web-based medical imaging ecosystem (e.g. https://github.com/cornerstonejs/react-cornerstone-viewport, https://github.com/dcmjs-org/, https://github.com/cornerstonejs/codecs).

Maintaining open source libraries requires a significant commitment of time and expense and we need to make strategic decisions to try to ensure that we can fund the projects long term. Until 2020, our primary source of funding was a National Cancer Institute grant to Dr. Gordon Harris at Mass General Hospital. We have been trying to apply for similar NCI grants since then but have not yet succeeded. For this type of grant, review panels focus largely on new capabilities delivered to the community and the evidence that this is having an impact on cancer research (and particularly U.S.-funded work). If we focus our resources solely on core libraries like Cornerstone, and not on higher-level user-facing tools like OHIF, we are not aiding our prospects for future NIH grant funding. We also need to choose where to allocate resources based on need. Cornerstone Core itself is currently stable enough to use without major problems, and so focusing on improving it without adding major new capabilities has very limited return on investment. There probably should have been releases more recently but I believe the automated release infrastructure is not set up for that repo.

Staffing is difficult because we do not have funding for even a single full-time employee (FTE) to work entirely on open-source, which means that we have to balance priorities with other projects (and these are the ones that pay the bills). Beyond that, the same people that write the code also need to help write the grants (which do not always receive funding, and then need to be rewritten, and have >9 month acceptance-to-funding lead times). There are very few grant types that are directly focused on funding open source, and we have been trying our best to get them.

The good news is that we recently received funding from the Chan-Zuckerberg Initiative for 2021 to work on improving our outreach and supporting the community (https://chanzuckerberg.com/eoss/proposals/enhancing-the-open-health-imaging-foundation-web-medical-imaging-framework/). The bad news is that it's only one year, and covers about half an FTE developer, so we are still on the "need to find funding" treadmill.

As part of our CZI funding we will be working on our communication with the community. We realize it is one of our weak points and aim to address it with proactive communication via a newsletter and striving to answer more questions on Github and the forums.

As you can imagine, we need to get a bit creative to fund open-source efforts. Along this line, we have a recently engaged academic-industry partnership project which will bring some exciting new capabilities to OHIF / Cornerstone, such as 3D tools and multi-modal fusion viewing. We have also contributed funding / time from other grant projects (e.g. https://reporter.nih.gov/project-details/10003193) to push forward some of our long-term goals. We are exploring internally at Mass General to understand how we can accept donations from the community to support more development. Another avenue to get fixes you need built is to engage a consultant (see https://docs.ohif.org/help.html#commercial-support).

If there are modifications or enhancements that you are looking to have implemented, would you consider providing support (either financial or in-kind) to help with development of the open source code base? 

In response to your queries, there are definitely some commercial alternatives if you are willing to pay for them (e.g. Laurel Bridge may have one? https://www.laurelbridge.com/). Another open-source alternative is Stone of Orthanc (https://www.orthanc-server.com/static.php?page=stone) which is very interesting, although it is AGPL licensed, so you may need to purchase a license exemption from the Osimis team.

Can you explain a bit more specifically about what your security concerns are? Stuff like dependabot alerts, etc?

Cheers,

Erik

[Michel Casabianca]

Hi Erik,

Thank you very much for this very interesting and documented answer. I guess it will be very useful for many Cornerstone users. I understand that you have priorities and won't release Cornerstone Core if this is not necessary.

I have no security concerns with this library, but if it was not maintained anymore (which is not the case obviously) this might be an issue. My only problem was the lack of documentation about some tools, but this is another story.

Best regards

[Marco Randon]

Dear Erik,

is there any roadmap regarding implementation of multi-modal fusion viewing in OHIF Viewer? I'm very interested in this topic.

Best regards,

Marco

[Erik Ziegler]

Hi Marco!

Here's a presentation we gave yesterday at NAMIC Project Week that has some rough information about timelines

https://docs.google.com/presentation/d/1KYNjuiI8lT1foQ4P9TGNV0lBhM6H-5KBs0wkYj4JJbk/edit#slide=id.p

Thanks,

Erik

--
You received this message because you are subscribed to a topic in the Google Groups "cornerstone platform" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/cornerstone-platform/BRP1d35Zyus/unsubscribe.
To unsubscribe from this group and all its topics, send an email to cornerstone-plat...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/cornerstone-platform/1efdeb4f-56cb-466c-b852-a1269b333f39n%40googlegroups.com.