Virchow: A Million-Slide Digital Pathology Foundation Model

• Main Authors: Vorontsov, Eugene, Bozkurt, Alican, Casson, Adam, Shaikovski, George, Zelechowski, Michal, Liu, Siqi, Severson, Kristen, Zimmermann, Eric, Hall, James, Tenenholtz, Neil, Fusi, Nicolo, Mathieu, Philippe, van Eck, Alexander, Lee, Donghun, Viret, Julian, Robert, Eric, Wang, Yi Kan, Kunz, Jeremy D, Lee, Matthew C. H, Bernhard, Jan, Godrich, Ran A, Oakley, Gerard, Millar, Ewan, Hanna, Matthew, Retamero, Juan, Moye, William A, Yousfi, Razik, Kanan, Christopher, Klimstra, David, Rothrock, Brandon, Fuchs, Thomas J
• Format: Journal Article
• Language: English
• Published: 14-09-2023
• Subjects: Computer Science - Computer Vision and Pattern Recognition; Computer Science - Learning; Quantitative Biology - Tissues and Organs
• DOI: 10.48550/arxiv.2309.07778

The use of artificial intelligence to enable precision medicine and decision
support systems through the analysis of pathology images has the potential to
revolutionize the diagnosis and treatment of cancer. Such applications will
depend on models' abilities to capture the diverse patterns observed in
pathology images. To address this challenge, we present Virchow, a foundation
model for computational pathology. Using self-supervised learning empowered by
the DINOv2 algorithm, Virchow is a vision transformer model with 632 million
parameters trained on 1.5 million hematoxylin and eosin stained whole slide
images from diverse tissue and specimen types, which is orders of magnitude
more data than previous works. The Virchow model enables the development of a
pan-cancer detection system with 0.949 overall specimen-level AUC across 17
different cancer types, while also achieving 0.937 AUC on 7 rare cancer types.
The Virchow model sets the state-of-the-art on the internal and external image
tile level benchmarks and slide level biomarker prediction tasks. The gains in
performance highlight the importance of training on massive pathology image
datasets, suggesting scaling up the data and network architecture can improve
the accuracy for many high-impact computational pathology applications where
limited amounts of training data are available.