World's First Wearable-Powered System Predicts Acute Inflammation With 90% Sensitivity

Traditional medical practices are often reactive, identifying infections only after symptoms appear or through delayed testing methods such as PCR. However, early physiological changes frequently go unnoticed until critical health events occur. This delay hampers timely interventions that could improve outcomes and reduce healthcare burdens. Now, researchers have developed a wearable-powered artificial intelligence (AI) system that detects early immune signals before symptoms arise, achieving nearly 90% sensitivity in identifying acute inflammation.

The solution, developed by researchers at McGill University (Quebec, Canada), can detect inflammation before symptoms appear, opening the door to earlier intervention, while saving lives and reducing healthcare costs by preventing complications and hospitalizations. To simulate real-world infection, the team administered a live attenuated influenza vaccine to 55 healthy adults aged 18–59 and monitored them over 12 days. Participants wore three commercially available wearables—a smart ring, watch, and shirt— which tracked heart rate, heart rate variability, body temperature, respiratory rate, blood pressure, sleep quality, and physical activity. Alongside this, researchers collected self-reported symptoms via a smartphone app and performed repeated blood tests and PCR testing. Over 2 billion data points were collected and used to train AI algorithms. Of the ten models built, the one with the fewest features was selected for real-world development, offering a practical and effective monitoring system.

The models based on wearable data significantly outperformed those relying solely on symptom reports, as many participants with inflammation showed no noticeable symptoms and vice versa. The results, published in The Lancet Digital Health, show that the platform even flagged SARS-CoV-2 infections before PCR tests confirmed them. The study's success has led to the creation of a startup to commercialize the platform for broader applications in detecting inflammation from viruses like rhinovirus, RSV, or SARS-CoV-2 using only wearable devices. The team now plans further clinical validation studies and to bring the technology into everyday clinical and personal health monitoring.

“By the time an infection is detected based on clinical symptoms or PCR testing, it is generally already well underway,” said Dennis Jensen, PhD, senior author of the study. "By enabling rapid, personalized, and objective early warning of systemic inflammatory events due to viral respiratory infections, our predictive tool gives patients and healthcare providers the chance to intervene early before critical health events occur.”