Page develops Pan-Tissue AI-enabled cancer detection technology

Digital pathology company Paige has announced that it has developed an application using artificial intelligence (AI) that can detect cancer from more than 17 different tissue types, including skin, lung and gastrointestinal tract, along with numerous rare tumor types and metastatic deposits.

The new app was developed using the company’s core model, called Virchow, which allows the technology’s developers to use more than four million digitized slides to train their model on multiple types of cancer. This is significantly different from the traditional approach of developing AI models that are trained to detect one type of tissue at a time, which can take months or years to build. The company says this is the first clinical application of AI-based cancer diagnosis.

“The early success of our foundational model was made possible by the size, quality and diversity of the datasets we used to build it,” said Siqi Liu, PhD, director of AI Science at Paige. “Page has access to one of the largest and most highly regarded pathology datasets worldwide, allowing us to use cutting-edge deep learning approaches to train systems to detect common, complex and even very rare cancers. The development of PAGE provides the pathology community with the most powerful tools for diagnosis, prognosis, biomarker development and targeted patient selection for precision therapy,” he continued.

According to Andy Moy, CEO of Paige, the company is focused on developing clinical digital pathology and AI-enabled technologies and will therefore seek FDA regulatory oversight for any of its products that are based on the Foundation Model technology and noted: “ We believe that FDA clearance is critical to ensure that regulatory and safety standards are maintained in the application of AI in cancer diagnosis across tumor types.

Paige was the first company to receive marketing approval from the FDA for AI digital pathology diagnostics for Paige Prostate, received in September 2021. Data from this study showed that the tool improved diagnostic cancer detection by 7.3% compared to the estimate of pathologist while reducing false negative diagnoses by 70% and false positives by 24%. The company is a 2018 spin-off from Memorial Sloan Kettering and initially used digitized slides from the cancer center, but has now attracted more than 1,000 other institutions in 45 countries to continue building on its more than four million digitized slides of cancer specimens .

The new pan-fabric application was created by Page with Microsoft Research through a collaboration announced last September. The goal is to capture much more subtle signatures that reflect the complexity of cancer and provide more accurate, individualized diagnoses.

By developing this app with Virchow’s help, Page said its ability to detect cancer using a variety of data sets has broader implications than diagnostics.

“In addition to its multi-tissue capabilities, the foundation model and use of its embeddings can also be used as a critical building block in a variety of upstream and downstream applications across the healthcare continuum,” said Raziq Yousfi, senior vice president of technology at Page . “By combining the results of the foundation model with data types from other modalities, including genomic, radiological and other health data, one can derive exponentially greater insights into the nature of cancer, its behavior and response to specific treatments.”

Leave a Comment

Your email address will not be published. Required fields are marked *