Paige Introduces Groundbreaking AI Transforming Cancer Detection Across Diverse Tissue Categories

Tuesday, January 09, 2024

Paige, a leading provider of end-to-end digital pathology solutions and clinical AI applications, has introduced a groundbreaking product called Virchow, developed through its Pathology Foundation Model. Utilizing a vast library of digitized images and computational resources from Microsoft Research, this innovative application is the first of its kind to detect cancer across more than 17 different tissue types, including skin, lung, and the gastrointestinal tract, as well as multiple rare tumor types and metastatic deposits.

Traditionally, the development of AI applications for pathology cancer detection required extensive datasets for individual tissue types, a process that often took months or years to achieve clinical-grade results. Paige has overcome these challenges by leveraging its unique Foundation Model. With data sourced from over 4 million digitized slides, Paige's approach removes the constraints of developing single tissue products, allowing for the efficient creation of AI applications for cancer detection across a wide array of tumor types—a significant advancement in AI-based cancer diagnosis.

Drawing from one of the world's largest and highly regarded pathology datasets, Paige employs state-of-the-art deep-learning techniques to train systems for detecting common, complex, and rare cancer entities. The resulting multi-cancer application is recognized as cutting-edge in cancer pathology AI. Paige is committed to ensuring the clinical-grade quality of its AI applications and intends to seek FDA regulatory oversight for products based on the Foundation Model technology.

By integrating outputs from the Foundation Model with various data types from genomics, radiology, and other health data, Paige aims to gain significantly greater insights into the nature of cancer, its behavior, and responses to specific treatments. This development is poised to provide the pathology community with robust tools for diagnosis, prognosis, biomarker development, and the targeted selection of patients for precision therapy.





Harvard Medical School - Leadership in Medicine Southeast Asia47th IHF World Hospital Congress