Pandemic Fighters

The Rapid Assessment of Threat Exposure (RATE) project is an infectious disease detection program by the Department of Defense (DoD).

The RATE project employs a supervised machine-learning algorithm trained by using hospital-acquired data from over 6,000 COVID-19 cases.

“Our algorithm is device agnostic. We can use biomarker data from any commercial grade, off-the-shelf wearable,” said Navin Natoewal, head of integrated technology solutions at Philips, one of the DoD’s corporate partners in this venture.

In this case, while Oura and Garmin products also participated, the most relevant component was the software employed, leaving the hardware open to the choice of each warfighter.

This algorithm caught COVID-19 cases 2.3 days prior to diagnostic testing, in some instances predicting cases six days in advance, according to a research paper. These diagnoses achieved 82% accuracy, with 11% rate of false positives.

The original software was developed by Philips under a Defense Threat Reduction Agency and Defense Innovation Unit-sponsored program.
Data employed has continued to grow, and the DoD plans to expand it beyond COVID-19.