Microsoft is working to train an Artificial Intelligence to be able to help detect cancer cases early on
Technology is making leaps and bounds in the world of medicine and early cancer detection is not far behind. Microsoft has announced an exciting collaboration with digital pathology provider Paige to develop the world’s largest image-based artificial intelligence model, and its goal is clear: to help detect cancer early.
This innovative artificial intelligence model is being trained on an unprecedented amount of data, including billions of images. Its ability to identify both common cancers and those rarer, notoriously difficult to diagnose promises to be an invaluable tool for physicians. At a time when medical staffing shortages and the number of cancer cases continue to rise, this collaboration could make a difference.
Latest technology
Pathologists, who play a crucial role in medicine, have been using a tried-and-true method to diagnose cancer for decades: examining tissue on a glass slide under a microscope. However, Paige is working to modernize this process. Its FullFocus visualization tool, approved by the Food and Drug Administration (FDA), allows pathologists to examine scanned digital slides on a screen instead of relying on a microscope.
A major challenge in digital pathology is the cost associated with data storage and collection. Digitizing a single slide can require more than a gigabyte of storage, which can be prohibitive for smaller healthcare systems. Paige, with its wealth of data and in collaboration with Microsoft, is overcoming these barriers.
AI to detect cancer
Paige’s original AI model used more than a billion images to identify specific cancers. But, the model under development with Microsoft is “orders of magnitude larger than anything out there.” It is being trained on 4 million slides to identify common and rare cancers, and is claimed to be the largest publicly announced computer vision model.
While the technology is powerful, it is not intended to replace pathologists. Instead, it is seen as a tool that will enrich their diagnostic capabilities. Microsoft’s Desney Tan emphasizes that AI should be driven by humans, not replace them.
This exciting project doesn’t stop here. Paige and Microsoft will publish a paper on the model, but the review and approval process will take time. However, the impact promises to be revolutionary. It will help solve storage problems in healthcare systems and speed up diagnostics. For many patients, it could mean the difference between waiting days or weeks to know their health status.