Chan Zuckerberg Biohub Network’s Post

🔬Revolutionizing pathology with #AI: researchers have developed a “human-in-the-loop” framework called https://lnkd.in/eTNfcBeW to overcome limitations of off-the-shelf AI models for pathology. They found it to be better and faster than humans or AI working alone. Led by #CZBiohubSF Investigator James Zou and Zhi Huang of Stanford University School of Medicine, the study validated the model in two use cases: examining histology images for colorectal cancer and endometriosis. By allowing humans to give real-time feedback to the AI model, clinicians were able to double their sensitivity in identifying plasma cells in endometrial biopsies while having a tool adapted for their needs. Read the paper in Nature Biomedical Engineering ⤵️ https://lnkd.in/e39ePE55

  • No alternative text description for this image
Dr. Suresh Kondeti

Postdoctoral Fellow specializing in scRNAseq and animal experiments at UNMC

2w

Impressive work! I am curious to know if this model is capable of counting the number of different cells and clustering them based on their specific features? Thank you.

Like
Reply
🤖 🧠 Thibault GEOUI 🧬 💊

Head of AI/ML for Drug R&D 💡- I reimagine 🧠 life sciences ❤️ and drug R&D 💊 with (FAIR) data 📊 and (AI) 🤖 technology

3w

Lise Bertrand Loïc Spica - you might find this interesting!

See more comments

To view or add a comment, sign in

Explore topics