Soham Mandal
sohmandal.bsky.social
Soham Mandal
@sohmandal.bsky.social
Data Scientist @ICR/PhD @University of Cambridge, Christ's College, EMBL-EBI. Does computer vision.
Super excited to see this one out! 🎉

Classpose is the result of my collaboration with @josegcpa.bsky.social to make large-scale, high-precision cell phenotyping both accessible and scalable.
🧬 Me + @sohmandal.bsky.social built an AI tool to automatically find and label all cells in huge microscope images: Classpose.

@trevorgraham.bsky.social @champalimaudf.bsky.social @champalimaudr.bsky.social @icr.ac.uk
December 23, 2025 at 6:55 PM
Reposted by Soham Mandal
All I want for xmas is... combined cell segmentation and phenotyping with high accuracy from whole slide histology images with #Classpose (semantic segmentation). www.biorxiv.org/content/10.6...

From the inimitable @sohmandal.bsky.social @icr.ac.uk
Classpose: foundation model-driven whole slide image-scale cell phenotyping in H&E
Cell segmentation and phenotyping in histopathology samples are essential techniques applied across diagnostic and research workflows. However, annotation by human experts requires significant time an...
www.biorxiv.org
December 22, 2025 at 12:53 PM
Reposted by Soham Mandal
Are you an #AI #digitalpathology developer struggling to make sense of all the pathology lingo? Don't worry because @sohmandal.bsky.social & @pathologop.bsky.social are here to help with their "Histopathology glossary for AI devs" journals.plos.org/ploscompbiol... published today - congrats both
The tumour histopathology “glossary” for AI developers
The applications of artificial intelligence (AI) and deep learning (DL) are leading to significant advances in cancer research, particularly in analysing histopathology images for prognostic and treat...
journals.plos.org
January 23, 2025 at 7:22 PM