Google announced in a blog post today that it's developed a prototype for a microscope system that uses artificial intelligence to automatically detect anomalies resembling cancerous growths in tissue samples. Google is calling it the Augmented Reality Microscope (ARM) platform, and it's currently capable of identifying breast and prostate cancer, although Google says it will be able to find any other type of cancer once its model is trained for it.
ARM can be integrated into light microscopes—even existing ones, Google says—and searches the field of view for areas that look like cancerous growths. When it finds one, the area is visually highlighted on an AR display under the eyepiece. The display is continuously refreshed as the slide is repositioned.
The process still requires trained professionals to operate the equipment, and models are currently trained on what to look for manually by pathologists. ARM could reduce diagnosis times, leading to faster treatment, and potentially catch things human eyes missed. You can read more about the project at Google's Research Blog, or if you're the technical type, check out the paper Google presented on ARM at the Annual Meeting of the American Association for Cancer Research today.