The Department of Defense has partnered with Alphabet Inc’s GOOG GOOGL Google to create an Artificial Intelligence-powered microscope that can assist doctors in identifying cancer.
This technology is called an Augmented Reality Microscope (ARM) and it is still in the early stages of development. Although it is not currently being used for patient diagnoses, initial research shows promising results.
There are currently 13 ARMs, one of which is located at a Mitre facility near Washington, D.C. Mitre, a nonprofit organization that collaborates with government agencies on technological challenges, is studying the ARM’s potential vulnerabilities in clinical settings.
Experts believe that the ARM could assist doctors in smaller laboratories dealing with staff shortages and increasing workloads.
Also Read: Apple Taps Seasoned Executive To Head ‘Secret’ Team Developing Glucose Monitoring Device – Stocks To Watch.
At first glance, the ARM looks like a regular microscope found in high school biology classrooms. It has a beige design with a large eyepiece and a slide tray for traditional glass slides.
However, it is connected to a powerful computer tower that houses AI models.
When a prepared glass slide is placed under the microscope, the AI can identify the location of the cancer by outlining it as a bright green line visible through the eyepiece and on a separate monitor.
The AI also evaluates the severity of the cancer and generates a black-and-white heat map on the monitor that shows the boundaries of the cancer in a pixelated format.
According to officials cited by CNBC, this technology could be a valuable resource for pathologists who do not have easy access to second opinions.
The cost of an ARM for healthcare systems typically ranges from $90,000 to $100,000.
The Defense Innovation Unit (DIU) of the Department of Defense recognized the potential of this technology and sought commercial partners to develop and test the ARM. Jenoptik was selected for the hardware, while Google was chosen for the software and ensuring the privacy and security of the data used to train the AI models.
