Orbit Image Analysis version 3.60 released

We are happy to announce that we just released version 3.60 of Orbit Image Analysis, a cross-platform research tool for Whole Slide Image Analysis. For details of research applications, see the recently published article. Downloads for Windows, Mac and Linux can be found here.

Version 3.60 includes a new tool for semi-automating annotations of TMA type slides, and support for object segmentation using deep learning models.

We updated the OMERO server support, so now it is possible to connect to version 5.6.x. We strongly urge you to follow the instructions here to upgrade OMERO to this version or newer. If you need to use an older version of OMERO server, then you will need to use an older version of Orbit Image Analysis. Thanks to the OMERO team for their help and advice with making the upgrade.

Furthermore, we updated to version 6.4.0 of Bioformats, for the latest file format support. In addition we updated to version 2.0.23 of the native NDPI reader libraries from Hamamatsu.

We made some technical changes that are documented here. As an end user you should only notice that the icons look nicer on monitors with resolution greater than 1920×1080. If you are a developer looking to contribute to Orbit Image Analysis development, we hope that our improvements to the build.gradle scripts will help you.

We’d love to hear your feedback on this release, what you love, what kind of analysis you’d love to be able to do. In case you have any issues please open an issue on the Github page, or add a comment in the orbit channel on forum.image.sc.

Thanks,

Manuel and Jon

6 Likes

Dear Manuel and Jon,
thanks for this tool, looks very interesting. I am new to the idea of performing WSI and associated analysis. I had only heard about QuPath and learned about it more yesterday at the @Neubias webinar (perhaps an intro into Orbit IA in youtube might also help disseminate its use). I liked the possibility to bring @StarDist into QuPath for nuclei and (roundish) cell detection. Ease of extensibility is definitely something nice. One thing that I would love to see in this or another WIS tool is the possibility to have basic alignment tools. We work a lot with brain slices in fluorescence and having an integrated tool where to align our images with a reference map would be fantastic, even if done in a simple manner. We’ll try this along QuPath.

1 Like

If your alignment is rigid, have you looked at:

This is great, and I’m looking forward to the ability to deep learning for object segmentation. I’ve notice an odd, intermittent issue where the intensity of an image will suddenly change. The images become very dim, which I din’t notice in Orbit 3.15. Otherwise, I am very impressed.

Best,
Sanjay

Hi Sanjay,

Thanks for the feedback. We’ll be working on even more improvements to the deep learning segmentation shortly.

Regarding the dimming, are you able to reliably reproduce that behaviour? If you can describe what you’re doing when it happens I can look into it.

Thanks,

Jon

Hi Jon,
I appreciate your quick response. I put together a couple of slides to show the issue. Please let me know what other information you need.

Thanks again,
Sanjay

Troubleshootin dimness issue.pdf (4.74 MB)

Thanks Sanjay for the detailed explanation. I have an idea what might be going on. Will look into it and see about a fix.

Dear @Zippy,
thanks a lot for the nice report, from what I can see it seems to be a display bug. I created a ticket on github for this:


Let’s follow up there.
Regards,
Manuel

Dear all,
we just released an Orbit bugfix version 3.63. This fixes on Orbit startup crash if the OMERO config was missing.
Regards,
Manuel

1 Like

…and another release: 3.64!
Now with Omero/websockets support and hopefully a fix for the “invalid parameters” startup bug which appeared for some users in combination with Omero.

Regards,
Manuel

1 Like