Google has joined forces with the U.S. Department of Defense to create an innovative tool for the medical field, the Augmented Reality Microscope (ARM). ARM is a cutting-edge AI-powered microscope designed to assist pathologists in diagnosing cancer and streamlining their workflow without disrupting established processes.
This state-of-the-art microscope uses real-time machine learning to enhance visual specimen inspection, especially in cancer diagnostics. The technology employs convolutional neural networks (CNNs), a specialized deep learning model ideal for image processing tasks. By extracting features from images using convolutional filters, CNNs accurately detect and classify cancer, revolutionizing pathology diagnostics.
The ARM concept was detailed in a 2019 scientific paper, showcasing how AI-based information can be seamlessly integrated into the current view of a sample, providing real-time insights during routine workflows. Early models of the ARM resemble traditional microscopes often found in high school biology classrooms, but they are augmented by a connected computer housing the AI models.
The ARM's purpose is to augment rather than replace existing digital pathology systems. Pathologists can use this tool to capture screen grabs of slides, significantly reducing storage costs. Presently, there are 13 ARMs in existence, each costing health systems approximately $90,000 to $100,000.