The Department of Defense is launching a digital pathology pilot program that would use AI combined with augmented reality microscopes to help with cancer detection. For the project, the Defense Innovation Unit awarded a contract to Google Cloud to deliver augmented reality microscopes to DoD medical facilities and develop AI models to help with cancer detection.
The Predictive Health project is expected to run for two years, with an initial rollout starting at certain Defense Health Agency facilities and Veterans Affairs hospitals. The DoD’s Joint Artificial Intelligence Center will provide funding and technical expertise for the project.
Google will use de-identified medical images to train the AI to identify cancers, using open source machine learning platform TensorFlow and its healthcare API. The models will use information from both public and private datasets.
Paired with the planned augmented reality microscopes, the AI would overlay information to help pathologists identify cells that might be cancerous.
“To effectively treat cancer, speed and accuracy are critical,” Google Cloud Vice President of Global Public Sector Mike Daniels said in a news release. “We are partnering with DIU to provide our machine learning and artificial intelligence technology to help frontline healthcare practitioners learn about capabilities that can improve the lives of our military men and women and their families.”
The goal would be to reduce errors in outpatient diagnoses, allowing for the detection and treatment of cancer earlier.
If the initial rollout is a success, they plan to expand the system more broadly across the U.S. Military Health System.
Earlier this year, the DIU awarded a separate cybersecurity contract to Google Cloud. The organization within the DoD is tasked with accelerating the adoption of commercial technology throughout the military.