Coinspeaker
Google and DoD Join Efforts to Build AI-Powered Microscope
Google LLC and the US Department of Defense (DoD) have collaborated to build an AI-powered microscope that will be able to identify cancer cells. Called an Augmented Reality Microscope, or ARM, the tool has been already successfully tested under the supervision of pathologist Dr. Nadeem Zafar at a VA hospital in Seattle.
ARM looks like and works similarly to a traditional microscope, where samples are placed on a glass tray and observed through an eyepiece. But compared to traditional microscopes, ARM also employs computer vision algorithms that help catch cells as benign or cancerous. ARM will share images that later can be closely monitored for a deep analysis.
The DoD’s Defense Innovation Unit (DIU) has initially invested in this technology to aid pathologists working in military hospitals with limited staff. The DIU believes that the ARM will accelerate the diagnostic process and provide experts with a second opinion when diagnosing diseases.
Optics company Jenoptik is in charge of building the ARM hardware, while Google is working on the software. The initial research into ARM looks promising, and once implemented, the system will bring the pathology to a different level.
Dr Nadeem Zafar commented:
“AI is here, and it’s going to keep developing. The point is not to be afraid of these technologies, but to triage them to the best use for our medical and health care needs.”
The estimated cost of ARM for health systems is between $90,000 to $100,000. Notably, the Department of Defense and Google aim to make this technology more widely accessible, particularly for smaller health systems that may not have enough resources for full-scale digitization.
AI Revolutionizing Microscopy
Artificial intelligence has proved to be a game-changer in many fields, microscopy and image are not an exception. The work of intelligent microscopy begins when the scientist returns from an excursion and examines what they have brought back under the microscope in the laboratory. Besides, machine learning is now widely used to build tools in the realm of cancer detection and diagnosing, potentially catching tumors or lesions that doctors could miss.
Scientists have developed AI tools to aid screening tests for several kinds of cancer, including breast cancer. AI-based computer programs have been employed to help doctors interpret mammograms for more than 20 years, but research in this area is quickly evolving.
So far, there have been 13 prototypes of the ARM. One of them is being tested at the MITRE lab near Washington DC. Researchers have conducted tests to detect breast cancer and metastasis in lymph nodes using four different AI algorithms. Results published in the Journal of Pathology Informatics demonstrated an accuracy level of about 94% in classifying cancerous and benign cells.
