The Defense Innovation Unit (DIU), the U.S. Department of Defense’s (DoD’s) division for adopting emerging technologies into innovations for the military and national security, has partnered up with the ubiquitous Google Cloud. Perhaps surprisingly, the partnership’s goal is not limited to combatting cyber threats but instead aims to aid in cancer detection.
Google Cloud will develop an artificial intelligence (AI) system for digital pathology to be used for research within a small number of Defense Health Agency and Veteran Affairs facilities. Depending on the program’s success, the DoD hopes to broaden the platform’s reach to include other U.S. Military Health System hospitals and clinics.
“To effectively treat cancer, speed and accuracy are critical,” said Google Cloud’s Vice President of the Global Public Sector, Mike Daniels. “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 U.S. sees over 1.7 million new cancer cases per year and almost 600,000 deaths due to the disease. According to BMJ Quality and Safety, approximately 5% of outpatient diagnoses contain errors—a massive misdiagnosis to the tune of 12 million patients each year.
Google Cloud and the DoD will utilize the open-source TensorFlow framework to develop machine learning models that can drastically reduce diagnostic error. Google’s Healthcare API will store and organize patient data for the highest level of privacy and ease of provider accessibility. The 3D models of the human body that AI can provide will make it easier for healthcare workers to find and visualize anomalies and will also help patients gain an understanding of how treatment options and surgeries work.
Using augmented-reality microscopes and AI models, Google hopes to provide the DoD with a new wave of cancer-detection tools that decrease human error and improve the accuracy of cancer diagnosis, as well as monitor the spread. Funding for the project will be provided by the DoD’s Joint AI Center, which will also lend its technical expertise.