🥳Alrighty - end-of-year push to get the final version of #CellSeg3D out!

👩‍💻 pip install napari-cellseg3d==0.2.2
📚 biorxiv.org/content/10.1...

We extended benchmarking, new videos, extended text, and a demo on cFOS 👋

👀Proofs! Excited to see #CellSeg3D formally out soon!

Big congrats to first author Cyril Achard et al 🎉

New tool CellSeg3D, from the group of blu.ski/@epfl-brainmind professor Mackenzie Mathis @trackingactions.bsky.social enables self-supervised 3D cell segmentation in fluorescent microscopy,👇 labeling time & 👆quantitative analysis -
paper: elifesciences.org/articles/99848
news: tinyurl.com/y2hpmbb9

I can add why we did this 🤗 - we wanted a self-supervised way to quantify whole brain cfos data!

We added demo code for using CellSeg3D after #BrainReg here github.com/C-Achard/cel...

cc @sainsburywellcome.bsky.social - thanks for #BrainReg!

🥰 Self-supervised, label-free 3D cell imaging is here

Congrats to Cyril on this very cool first, first author publication 🏆! This started when he was a bachelor's student (and he's now completing his masters!).

And many thanks to my other awesome co-authors🙏

elifesciences.org/articles/99848

✍️ New in #eLife: #CellSeg3D introduces #WNet3D, a self-supervised 3D #segmentation method for #microscopy data — no labels needed. Claims to outperform #Cellpose/#StarDist on 4 datasets. Includes #opensource plugin (#Napari) + full 3D annotated #cortex dataset. Will test it later.
@napari.org

Video Demos here: youtube.com/playlist?lis...

All source code: github.com/AdaptiveMoto...

All Figure Notebooks: github.com/C-Achard/cel...

New 3D mesoSPIM GT data: zenodo.org/records/1109...

Pretrained Transformer and WNet3D models auto-downloadable on HuggingFace 🤗

Self-supervised, label-free 3D cell imaging is here

EPFL researchers introduce CellSeg3D, a self-supervised tool for 3D cell segmentation in fluorescence microscopy, eliminating the need for manual labeling and enhancing accessibility for various biological studies.
youtu.be/3UOvvpKxEAo

www.biorxiv.org/content/10.1...

Tested #CellSeg3D and #Cellpose on their example c5image dataset. Both segmentations look reasonable out-of-the-box, without any deep parameter tuning. With some extra effort, one could likely push either further I guess. Overall, both tools perform quite well on this small sample data set.

A paper using RRID:AB_2910652 from @thermofishersci.bsky.social was just published in a journal. We value the author's support of reproducibility. #STMpublishing #ReproducibleResearch #OpenScience

Tested both tools also on a more realistic 3D stack from the #ImageJ sample library. #Cellpose is fast and works well out of the box. #CellSeg3D is, however, slower and needs some further tuning for instance segmentation I guess — but looks promising! Definitely worth a try 👌 #BioimageAnalysis

Authors used C57BL/6J in their study. Including #RRIDs will make this less ambiguous.

SciScore made a table with this resource, see “Automated Services” module (download as csv, xml or #jats) #reproducibility #OpenScience

To wrap this up: Both tools are easy to test. I recommend trying them on your own data to see what works best for your use case.

I’ll include #CellSeg3D in our next #Napari #bioimage analysis course. Curious what impressions and feedback the students will share. 🧪🔍

CellSeg3D: self-supervised 3D cell segmentation for microscopy https://www.biorxiv.org/content/10.1101/2024.05.17.594691v1

CellSeg3D: self-supervised 3D cell segmentation for microscopy https://www.biorxiv.org/content/10.1101/2024.05.17.594691v1

CellSeg3D: self-supervised 3D cellsegmentation for fluorescence microscopy [updated]
Self-sup. models (WNet3D) for 3D cell seg. in microscopy, validated on diverse data, perf. similar to sup. methods.