Accurate predictive health monitoring could help tens of millions of Americans each year, saving them from pain, suffering, and even death. Early disease detection and treatment often heads off worsening symptoms, more dramatic treatment protocols, unwanted side effects, and related complications. In too many cases, however, patients and physicians alike end up saying something along the lines of “I wish we’d caught this earlier.”

Royal Philips works with two units of the U.S. Department of Defense, the Defense Threat Reduction Agency (DTRA) and the Defense Innovation Unit (DIU) on an early warning algorithm to detect infections before symptoms occur. The joint program has two goals: to improve individual readiness by reducing downtime and to contain the spread of communicable diseases.

Philips recently announced success with a prototype predictive health monitoring algorithm. The prototype was significantly more effective than chance at predicting infections up to more than 48 hours prior to the subjects’ experiencing observable symptoms. Philips analyzed data including 165 different biomarkers with more than 41,000 cases of hospital-acquired infection (HAI). The team also discovered that the combinations of vital signs and biomarkers with high predictive rates varied depending on the time span before clinical observations of infection. These two breakthroughs form the basis of the next steps.

The collaboration’s ultimate goal is a wearable device for soldiers that passively monitors health to predict potential infections. The technology could also be used in military and civilian hospitals to catch and treat impending infections.