By Diagnostics World Staff
August 16, 2024 | Last week, Paige, in collaboration with Microsoft, unveiled the second generation of Virchow, its million-slide foundation model for cancer. Virchow2 and Virchow2G offer a deeper understanding of cells and tissue, aiming to redefine cancer diagnosis and treatment.
Built with a diverse dataset of over 3 million pathology slides from over 800 labs and 45 countries, Virchow2 and Virchow2G offer unparalleled data diversity and depth. Trained using de-identified data from over 225,000 patients, these models encompass a broad spectrum of gender, race, ethnicity, and geographical regions, providing a more holistic understanding of cancer.
This comprehensive dataset also includes over 40 different tissue types stained with H&E and diverse immune-stains (IHC), making it suited to a wider variety of applications. With 1.8 billion parameters, tripling the size of previous models, Virchow2G is the largest pathology model ever created. Trained in collaboration with Microsoft researchers and using Microsoft’s advanced supercomputing infrastructure, these models set a new record in AI training scale, surpassing previous performance standards according to a report published on arXiv.
“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,” said Thomas Fuchs, Dr.Sc., Founder and Chief Scientist of Paige in a press release.
“This second generation of Paige’s Virchow model outperforms anything in the industry and continues to grow in knowledge and capability, bringing us closer to making precision medicine a reality,” added Razik Yousfi, Senior Vice President of Technology at Paige, in the same statement. “We are not only expanding capabilities, increasing accuracy, and reducing time in the cancer diagnosis process, but also pushing the boundaries of what’s possible. Our goal is to bring continue to bring the most advanced AI to pathology, leading to better patient outcomes and significant advancements in disease understanding and treatment.”
Virchow Foundation Model Technology and its Impact on Cancer Today
To aid in the detection of cancer, Paige has used its foundation model technology to develop a universal clinical AI application for pathologists to aid in the identification and diagnosis of cancer across over 40 tissue types. Diagnosis can often be time-consuming and error prone, but AI can help the pathologist identify even rare cancers rapidly and accurately highlight each area of concern for further review by the pathologist, making it easier and more efficient for them to assess and reach a diagnosis.
Beyond cancer detection, and to better understand the genetic markers of cancer, Paige has also developed AI modules that serve as pre-built solutions for life sciences, pharmaceutical companies, and research entities. These modules enable precise therapeutic targeting, novel biomarker identification, and optimized clinical trial design.
The result is more successful clinical trials, faster time-to-market for new therapies, and significantly enhanced R&D pipelines for pharmaceutical and life sciences organizations. By integrating these advanced AI capabilities, life sciences organizations can enhance research efficiency, uncover new insights, and drive innovation across various scientific disciplines.
These AI modules, which include Pan Cancer Detection, Pan Cancer Digital Biomarker Panel, and Cellular Analytics, along with Virchow2 and Virchow2G, are now available for commercial use. Virchow2 also joins Paige’s OpenPFM Suite on Hugging Face for non-commercial research purposes.