This is tremendously exciting! The largest #OpenSource initiative in #ComputationalPathology by far. It includes the #Virchow #FoundationModel, trained on billions of images from 1.5 million slides, available under the Apache 2.0 license for everyone to use. On top of that, the #PRISM model, which provides whole-slide embeddings for pan-cancer detection, subtyping, and IHC biomarker prediction. Download them and try them out! It takes an entire field to revolutionize cancer research and make a tangible clinical impact on patient care.
Our open-source initiative, Open Paige Foundation Model (Open PFM) provides access to our best-in-class technology and underscores our commitment to driving innovation, collaboration, and accessibility in AI research. 𝗖𝘂𝗿𝗿𝗲𝗻𝘁 𝗺𝗼𝗱𝗲𝗹𝘀 𝗮𝘃𝗮𝗶𝗹𝗮𝗯𝗹𝗲 𝗼𝗻 𝗢𝗽𝗲𝗻 𝗣𝗙𝗠: 𝗩𝗶𝗿𝗰𝗵𝗼𝘄: • 1.5 million WSI, 600 million parameters • State-of-the-art performance published in Nature Medicine • Pan-cancer and rare cancer detection • Tile Embeddings & fine-tuning capabilities • Available under the Apache 2.0 License 𝗩𝗶𝗿𝗰𝗵𝗼𝘄 𝗣𝗲𝗿𝗳𝗼𝗿𝗺𝗮𝗻𝗰𝗲: https://lnkd.in/gSDiJisg 𝗠𝗼𝗱𝗲𝗹 𝗔𝗰𝗰𝗲𝘀𝘀: https://lnkd.in/eMFmpRpS 𝗣𝗥𝗜𝗦𝗠: • Built using Virchow • 587,000 additional WSI, and 195,000 clinical reports • Pan-cancer detection, subtyping and IHC biomarkers prediction • Slide Embeddings & fine-tuning • Available under the CC-BY-NC-ND 4.0 License 𝗣𝗥𝗜𝗦𝗠 𝗣𝗲𝗿𝗳𝗼𝗿𝗺𝗮𝗻𝗰𝗲: https://lnkd.in/e482uPMP 𝗠𝗼𝗱𝗲𝗹 𝗔𝗰𝗰𝗲𝘀𝘀: https://lnkd.in/e27vM59V 𝗚𝗶𝘁𝗛𝘂𝗯: https://lnkd.in/e5JQqEQe All available models of Open PFM can be accessed on Hugging Face: https://lnkd.in/e_RemKce
This is great! Thanks for making this open source.
Assistant Research Professor at New York University
4moThis is incredibly exciting and puts Paige at the forefront of AI in Pathology research. Kudos!