DoD, Philips partner for AI-enabled infection control

The technology examines vital signs and other biomarkers to predict the likelihood of infection up to 48 hours ahead of clinical suspicion.
By Max Sullivan
01:46 PM

A new technology based on artificial intelligence is being used to identify infections up to two days before symptoms appear, intended for U.S. service members but expected to be used widely by civilians. 

WHY IT MATTERS
The technology, developed by Royal Philips and the U.S. Department of Defense, uses AI to look at vital signs and other biomarkers to predict the likelihood of infection up to 48 hours ahead of clinical suspicion. The healthcare technology company and DoD collaborated on an 18-month project aimed at detecting infections in patients before they show signs or symptoms, announcing their results this week.

Philips and the DoD said the technology will be broadly applicable to civilian healthcare settings. They said it will eventually be used in scenarios where vital signs and biomarkers fluctuate, like during physical exertion and heat stress.

THE LARGER TREND
Philips and DoD intend to use the technology in devices worn by troops to monitor their health in a non-invasive way. Their hope is to detect infections among troops before overt symptoms show, at which point the infected have often already become sick and been exposed to others.

The project, called Rapid Analysis of Threat Exposure, or RATE, revealed how combinations of significant vital signs and biomarkers varied based on time before clinical suspicion of a hospital acquired infection, according to Philips.

The company and DoD called RATE the first large-scale look into predicting pre-symptomatic infections in humans, conducted as part of an overall effort to improve readiness.

The technology requires a large dataset that draws from more than 41,000 cases of infections, according to Philips. The dataset was extracted from a large repository of more than 7 million hospital patient encounters. The pared down cases were used as a surrogate dataset for infection in otherwise healthy military personnel and analyzed to develop a predictive algorithm of disease.

Philips said RATE's algorithm to predict infection 48 hours before clinical suspicion "can be characterized technically as area-under-the-curve of 0.853." 

For comparison, the performance lies in between blood-based breast and prostate cancer screening tests and an enzyme immunoassay based first-tier Lyme disease test.

ON THE RECORD
"The unique capability that Philips has produced enables the chemical and biological defense medical paradigm to shift from a reactionary focused one to a predictive one," said Edward Argenta, Science and Technology Manager for the Joint Science & Technology Office at the Defense Threat Reduction Agency.

"This provides our commanders with insight into their troops’ future readiness levels and can influence mission planning and overall military effectiveness," he added.

"By coupling large-scale data, with our experience in AI and remote patient monitoring with DTRA’s drive for innovation, we were able to develop a highly predictive early-warning algorithm based on non-invasively collected biomarkers," said Dr. Joseph Frassica, chief medical officer and head of research for Philips North America, in a statement. "These results can be extended in future work to also apply to other healthcare settings."

Max Sullivan is a freelance writer and reporter who, in addition to writing about healthcare, has covered business stories, municipal government, education and crime.

Twitter: @maxsullivanlive
Email: maxesullivan@gmail.com

Healthcare IT News is a HIMSS Media publication.