Microsoft has partnered with digital pathology provider Paige to develop the largest image-based artificial intelligence model for cancer identification. The AI model is being trained on a vast amount of data, including billions of images, enabling it to identify both common and rare cancers. The aim is to support doctors who are struggling with staffing shortages and increasing caseloads. Paige specializes in digital and AI-powered solutions for pathologists, who play a crucial role in diagnosing patients and determining treatment plans. By digitizing the pathologists’ workflow, Paige aims to improve accuracy and efficiency within the field.
Paige has already received FDA approval for its viewing tool FullFocus, which allows pathologists to examine digital slides on a screen instead of a microscope. The company has also built an AI model to assist pathologists in identifying breast, colon, and prostate cancer. However, the costs associated with data storage have posed a challenge for smaller health systems. Paige has significant amounts of data due to its association with the Memorial Sloan Kettering Cancer Center, allowing it to develop its own AI-powered solutions. To further expand its operations and develop a tool that can identify more cancer types, Paige sought assistance from Microsoft. Microsoft’s cloud storage and infrastructure have helped Paige build a larger, more advanced AI model than anything available currently.
Paige’s new model, training on 4 million slides, is the largest computer vision model ever publicly announced. The partnership aims to enrich pathologists with AI tools rather than replacing them. While the model has shown promising results, there is still a long journey ahead, including peer review and collaboration with regulators. However, the AI model has the potential to solve storage issues for health systems and improve diagnostic efficiency, particularly in overwhelmed community clinics. Ultimately, the goal is to democratize access to healthcare.