How do organelles change over the course of a stem cell’s life? Allen Distinguished Investigators @cohenlaboratory, @AssafZaritsky, and @mattersOfLight are using AI and cell engineering to explore the #FrontierScience of stem cell differentiation. 🔗 https://t.co/pq6AeT7loq

We have a fun update on DynaCLR ( https://t.co/xcCmu0bURH), a self-supervised method for learning cell state dynamics from 3D multi-channel movies! Congratulations, team @czbiohub, for reaching this milestone! Special thanks to Carolina Arias and her team for the close

Virtual staining offers a promising approach to map dynamic cellular systems across various cell types + conditions, broadening the scope of live-cell imaging. Congrats to Shalin Mehta & the team at #CZBiohubSF 🎉🙌 https://t.co/kfxdF31TDV

🚫 No dyes. No bleaching. 🔬 Just AI + label-free microscopy = vivid virtually stained images New in @NatMachIntell: A deep learning model that enables robust virtual staining across microscopes, cell types & conditions. #CZBiohubSF @mattersOfLight explains:

I am excited to join the next cohort of Allen Distinguished Investigators and looking forward to parsing the role of organelle interactions with imaging and AI in partnership with @cohenlaboratory and @AssafZaritsky!

How do organelles change over the course of a stem cell’s life? @cohenlaboratory, @AssafZaritsky, and @mattersOfLight are using machine learning and cell engineering to study the impact of organelle-organelle interactions on stem cell differentiation. #FrontierScience

Writer @Fairy__Hedgehog covers how microscopy researchers use AI, including @mattersOfLight, who has pioneered AI-augmented label-free light microscopy for biology at #CZBiohubSF, and #CZBiohubSF Investigator @optrickster, who uses AI for lensless imaging.

Thanks to students and faculty for their company and contributions. Hope to see many of you again soon!

Take a look at course repository for tutorials on the state-of-the-art deep learning methods for bioimage analysis:

Directing this year's DL@MBL course with @janfunkey, @sagzehn, @ilastik_team, and our awesome TAs was one of my most intense and fulfilling projects of 2024. The spirits of curiosity, collaboration, and Captain Kidd 🙃 made the course so much fun! https://t.co/X5yb1nciC7

The combination of virtual staining and ultrack enables label-free segmentation and tracking: https://t.co/RPoupikQSl This has already enabled gentler imaging and analysis of cell dynamics in our projects.

Shout out to @jobragantini and @loicaroyer for leading this effort to develop a versatile cell tracking method. Check it out!

1/ 📡Exciting news! 🚀 Our #Omega paper has been accepted at @NatureMethods! Omega is a @napari_imaging plugin that uses Large Language Models to assist users with bioimage processing and analysis tasks! Check out our original project announcement (see below), read the paper:

I've been so out of breath trying to deal with grant deadlines that I've barely had time to pause and appreciate the massive amount of progress we made in the last year in spite of the move from Salk to UCSD and hanging around in temporary space (permanent lab is done 7/31!) 🧵/1

Finally, thanks to Priscilla Chan and Mark Zuckerberg for their support, and to the leadership at @czbiohub for creating an environment where partnerships between technologists and biologists are the norm. /end

Successful execution required the integration of ideas across optics, algorithms, automation, high-performance computing, cell biology, and neuropathology. I am thankful to my colleagues for their expertise and for placing this bet with me. 14/n

@czbiohub @LiHao_Yeh @orctweets This is one of those projects that has taken several iterations and years to finish - look at the timestamps! Special thanks to @rita_strack for being an amazing editor throughout this process. 13/n

@czbiohub @LiHao_Yeh We needed a calibration target to evaluate the measurements of 3D dry mass and 3D orientation made with PTI. In addition to biological structures of known architecture and isotropic beads, we used anisotropic glass targets built by the Peter Kazansky lab @orctweets. Intriguing

@czbiohub As I mentioned before, the design complexity lies in the algorithms and calibration. @LiHao_Yeh validated the wave optical model of the microscope and the reconstruction algorithm with methodical simulations of isotropic beads and an anisotropic patterns. 🌊💡 11/n

@czbiohub PTI is easy to multiplex with fluorescence and H&E imaging, allowing label-agnostic mapping of cell and tissue architecture. In this movie, we image a standard H&E stained cardiac tissue slice. 10/n