Cambridge, Mass. – Royal Philips (NYSE: PHG, AEX: PHIA), a global leader in health technology, the U.S. Department of Defense’s (DoD) Defense Threat Reduction Agency (DTRA), and Texas A&M University’s School of Education and Human Development, its College of Engineering, its School of Medicine, and the Texas A&M Engineering Experiment Station (TEES), today announced the Persistent Readiness through Early Prediction (PREP) clinical trial, aimed at understanding the body’s physiological response to different infections such as pneumococcal pneumonia (respiratory) and typhus (non-respiratory). Philips researchers will analyze the trial data and use it to further train Artificial Intelligence (AI) to differentiate between and predict types of infections.
This work builds upon previous research on COVID-19 detection using Philips’ Rapid Analysis of Threat Exposure (RATE) algorithm in a prospective field study involving over 10,000 DoD participants that was recently published in Scientific Reports. The research found that combining commercial off-the-shelf (COTS) wearables with the predictive power of the RATE algorithm can effectively predict COVID-19 exposure days before diagnostic testing. Through the PREP clinical trial, the RATE algorithm’s risk scoring could potentially be expanded to discriminate between multiple types of infections. By expanding it to include common respiratory infections like pneumococcal pneumonia, RATE can help further fine-tune infection scores to distinguish between these diseases.
While infectious diseases can threaten the health and productivity of all aspects of society, it is particularly troublesome for organizations that provide essential services. Philips has been collaborating with Texas A&M’s Center for Translational Research in Aging & Longevity (CTRAL), the School of Medicine, TEES Center for Remote Health Technologies and Systems (CRHTS), and TEES Center for Applied Technology (TCAT) to implement the infrastructure, hardware, and software logistics required to complete the clinical study.
Participants in the clinical study are provided with a variety of COTS wearable devices to monitor their daily response for four weeks. During this time, the participants visit CTRAL on six separate occasions to complete a variety of measurements onsite in their clinic. Measurements include standard vitals monitoring using Philips patient monitoring IntelliVue system, onsite audio and photo recording, questionnaires, and biomarkers from breath and blood. After the study concludes, the collected data will be fed into the Philips machine learning (ML) pipeline to expand the RATE capability to predict infection as well as the type of infection beyond just COVID-19.