Paige Launches AI Models for Cancer Diagnosis — Built With 3 Mn+ Path Slides from 45 Countries

Paige, a global leader in clinical AI applications for cancer, has collaborated with Microsoft to unveil the second generation of Virchow, its foundational AI model for cancer. As the newest additions to Paige’s repository of foundation models, Virchow2 and Virchow2G are designed to redefine cancer diagnosis and treatment.

The models are built with a dataset of over 3 million pathology slides from over 800 labs in 45 countries. Trained using de-identified data from over 225,000 patients, Virchow2 and Virchow2G envelope a wide spectrum of gender, race, ethnicity, and geographical regions, providing a more holistic understanding of cancer.

According to an official update, the dataset also contains over 40 different tissue types stained with H&E and diverse immune stains (IHC), rendering it fit for a broad range of uses.

Paige claims that Virchow2G is the largest pathology model to date, with an impressive 1.8 billion parameters. The model was trained in partnership with Microsoft researchers and by utilizing Microsoft’s advanced supercomputing infrastructure, setting a new record in AI training scale.

“Our collaboration with Microsoft has been pivotal in the development of Virchow, which has already earned recognition in Nature Medicine,” said Thomas Fuchs, Dr.Sc., Founder and Chief Scientist of Paige. “We are merely scratching the surface of what these foundation models can achieve in transforming our understanding of cancer through computational pathology. Virchow’s immense scale unlocks key information that can be used to drive groundbreaking innovations, enabling precise diagnostics, targeted treatments, and personalized patient care. This is the beginning of a new era in oncology, where technology and science converge to combat cancer more effectively than ever before.”