Google will be working with the US Department of Defense to create a new artificial intelligence (AI) tool that will study and detect cancer among military veterans.
Through the Pentagon's Defense Innovation Unit (DIU), Google is set to develop a new AI model of information that can be integrated with virtual reality microscopes, according to a Fox News exclusive. Google's new machine learning model is expected to help doctors using microscopes in mapping out cancer, allowing the study of its distinct structure.
Improving Health Care for the Nation's Servicemen
"Health care is critical to the military's force readiness," said Mike Daniels, vice-president of Google Cloud's global public sector, in an interview with Fox News. "More important than that, or equally important, is funding for certain cancer-related programs." Daniels also claimed that the innovations rolled out in this program could lead to breakthroughs not only in cancer research but also in general health care.
The Defense Health Agency (DHA), the joint integrated support agency for the Army, Air Force, and Navy, allots $1.7 billion of its yearly budget for cancer research.
Daniels explained that there are two goals Google aims to achieve with the Pentagon program: (1) better patient outcomes in relation to diagnosis and (2) to create "some efficiency" in helping pathologists pour through the large volume of data they deal with.
"It's an important practical application of AI at the point of care, to drive real-world outcomes that matter for our country," the Google Cloud VP added.
Aside from working to reduce medical costs, the program also focuses on both active-duty servicemen and veterans through DHA treatment facilities and Veterans Affairs (VA) healthcare facilities across the country.
A Virtual Assistant for Pathologists
According to Daniels, there is a "limited set of military hospitals and VA facilities" equipped with their own augmented-reality microscopes, which has received the new Google AI model to serve as the pathologist's virtual assistant as they examine the slides.
Using an algorithm-relay match, the AI can inform the pathologist of the areas that require additional attention. Daniels noted that their responsibilities would include "day-to-day program management, architecture, and technical implementation of that aspect."
Aashima Gupta, Google Cloud's global health care solutions director, also told Fox in an interview that they cannot understate the level of relief this program will provide for practicing pathologists. With the Google AI model, the workload of pathologists will be reduced, allowing them to focus on their other undertakings such as research and publishing findings, instead of pouring through a large amount of data.
"Looking through a pathologist's microscope and examining the tumor in a patient, that's the gold standard today," Gupta said, noting that their task of looking into the tumor tissues is laborious and burdensome. Before diagnosing the stage of a patient's cancer, pathologists' have to analyze small cells from the sample. In this aspect, Google aims to limit the uncertainty involved by allowing the AI model to overlay and identify the areas that need additional attention.
The Google AI model will first focus on so-called "targeted cancers" such as prostate, colon, and breast cancers, lymph node metastasis, and cervical dysplasia. However, the program can also expand in the future to include other cancers and can also lead to further improvements in AI technology, which can be used for different applications.
Check out more news and information about Artificial Intelligence on Science Times.