After successful development and pilot, technology using biomarkers from wearables to predict infection is set to expand within the US Department of Defense.


RT’s Three Key Takeaways:

  1. Expanded Defense Investment: The US Department of Defense awarded Philips a $25 million contract to advance the rollout of the RATE AI algorithm, aimed at detecting infections early to enhance military personnel readiness.
  2. AI-Driven Health Monitoring: RATE uses AI and data from wearable devices to identify pre-symptomatic infections, potentially detecting illness days before symptoms appear, reducing downtime for service members.
  3. Four-Year Deployment Plan: The contract will support a four-year expansion of RATE, allowing more active-duty personnel access to predictive health monitoring, aiding in timely intervention and overall troop well-being.

Royal Philips announced it will accelerate the predictive health monitoring of the Rapid Analysis of Threat Exposure (RATE) algorithm to support the well-being and readiness of active-duty service personnel for the US Department of Defense. 

Over the next four years, the Joint Program Executive Office (JPEO) for Chemical, Biological, Radiological, and Nuclear Defense will invest $25 million to make the technology more accessible in the field.

RATE, in advanced development, is an early detector of pre-symptomatic infection in humans. It is part of efforts to improve the readiness monitoring of active-duty personnel and has broad application in healthcare settings. The algorithm was developed as an early warning system that can reduce an individual’s downtime and aid in quickly containing the spread of disease by allowing exposed individuals to self-isolate or seek medical attention sooner.

“Our mission is to protect our troops and ensure we can quickly detect and prevent the spread of illness that can impact their wellbeing and our ability to protect our nation,” says Lorenzo Hankla, JPEO wearables program officer, in a news release. “Giving them the option of using readily available off-the-shelf wearables that are trained on clinical data allows them to passively monitor their health and quickly act if their infection score is high. In the wake of the pandemic, our troops need access to new AI-enabled technologies that can have a meaningful impact on their readiness and ensure overall military operations.”

Evolutions of the RATE Algorithm

The first version of the RATE algorithm, developed in 2019, used large-scale data science, machine learning, and trade-space analyses across demographics and 163 different biomarkers (vital signs and laboratory measurements) from a Philips data set of over 36,000 cases of hospital-acquired infections to develop a risk score. The risk score worked for multiple general types of infection, including common respiratory infections like pneumococcal pneumonia. These results were published in a 2023 study in Frontiers in Medicine.

A wearable version of the algorithm, prompted by the COVID-19 pandemic, leveraged biometric data from commercial-off-the-shelf wearables to enable early detection of infection. A 2022 study published in Nature’s Scientific Reports showed how, in a cohort of 9,381 Department of Defense personnel, the algorithm was able to detect COVID-19 infection on average 2.3 days before symptoms appeared and up to six days prior. The study highlighted the utilization of algorithm-powered wearables to aid military readiness.

“Protecting the physical health and well-being of active service members is paramount,” says Jeff DiLullo, chief region leader of Philips North America and US Army Veteran, in a release. “Our military personnel have unique care needs depending on their role and environment, and thanks to AI-enabled, predictive innovations like RATE, we can enable more timely intervention and personalized care, reducing downtime and improving outcomes—not only for individuals but entire units, too. The faster we diagnose and treat, the sooner these brave men and women are ready for duty.”

ID 149438553 © Rafael Henrique | Dreamstime.com