They undergo intensive training and robust cycles of vaccinations: they have strong fiber, but they are still human beings, and sometimes they get sick. They are the soldiers, and when they are not well they can put critical operations at risk.
La DTRA (an acronym that stands for Defense Threat Reduction Agency) tries to overcome this problem with a predictive algorithm that can predict a person's illnesses due to different causes, from colds to exposure to biological agents. up to 48 hours before you start showing any symptoms.
“Think of it as a control system for the human body,” he says Edward Argenta, scientific and technological manager of the Joint Science and Technology Office of the DTRA .
DTRA has partnered with the US Defense Innovation Unit to leverage its special capability (that is, the possibility of placing orders outside of federal regulations). The "limitless" possibilities of this partnership led to research on the algorithm to predict diseases, started with the collaboration of Royal Philips, a company specializing in pharmaceutical technology.
Using its own globally collected datasets, Philips was able to develop a unique algorithm for the Department of Defense. Using 165 distinct biomarkers in 41.000 cases, the Philips team was able to create the algorithm called Rapid Analysis of Threat Exposure, or RATE, which according to the company can “predict infection 48 hours before clinical suspicion” with an accuracy better than 85%.
“For comparison, this accuracy rate is currently found in blood draw-based breast and prostate cancer screening tests and a first-tier Lyme disease test based on enzyme immunoassays.”, according to a company statement.
“By combining large-scale data, with our experience inartificial intelligence and remote patient monitoring with DTRA's commitment to innovation, we were able to develop a highly predictive early warning algorithm based on non-invasively collected biomarkers”Says Joe Frassica, medical director and research manager for Philips North America.
“Although the RATE data is derived from acute care settings, we believe it is adaptable to active duty personnel.”