Researchers at McGill University have developed an artificial intelligence platform capable of predicting respiratory tract infections before symptoms appear. This groundbreaking study, described as a "world first," involved participants wearing a smart ring, watch, and T-shirt equipped with sensors to monitor their biometric data.

The AI system analyzes this data to accurately predict acute systemic inflammation, an early indicator of respiratory infections like COVID-19. The findings, published in The Lancet Digital Health, suggest that this technology could enable doctors to address health issues earlier, particularly for vulnerable patients who may face serious risks from new infections. Additionally, it has the potential to lower healthcare costs by preventing complications and hospitalizations.

"We were very interested to see if physiological data measured using wearable sensors could be used to train an artificial intelligence system capable of detecting an infection or disease resulting from inflammation," said Prof. Dennis Jensen, the study's lead author from McGill's department of kinesiology and physical education. He emphasized the goal of detecting early physiological changes to predict illness onset.

Jensen noted that the AI model is the first to utilize physiological measures—such as heart rate, heart rate variability, body temperature, respiratory rate, and blood pressure—rather than relying solely on symptoms to identify health issues. Acute systemic inflammation is a natural defense mechanism that typically resolves itself but can lead to severe health complications, especially in individuals with pre-existing conditions.

"The whole idea is kind of like an iceberg," Jensen explained. "When the ice cracks the surface, that’s when you’re symptomatic, and then it’s too late to really do much to treat it."

In the study, researchers administered a weakened flu vaccine to 55 healthy adults to simulate infection. Participants were monitored for seven days before and five days after vaccination while wearing the smart devices. Researchers also collected blood samples for biomarkers of systemic inflammation and used PCR tests to identify respiratory pathogens, alongside a mobile app for symptom reporting. In total, over two billion data points were gathered to train machine learning algorithms.

The team developed ten different AI models but selected the one that required the least data for further use. This model successfully detected nearly 90% of actual positive cases, making it practical for daily monitoring. Jensen pointed out that no single data source from the ring, watch, or T-shirt was sensitive enough on its own to gauge the body’s response effectively.

"An increase in heart rate alone may only correspond to two beats per minute, which is not really clinically relevant," he said. "The decrease in heart rate variability can be very modest. The increase in temperature can be very modest. So the idea was that by looking at several different measurements, we would be able to identify subtle changes in physiology."

The algorithms also identified systemic inflammation in four participants who contracted COVID-19 during the study, flagging their immune responses up to 72 hours before symptoms emerged or infections were confirmed by PCR testing.

Ultimately, the researchers aim to create a system that alerts patients to potential inflammation, prompting them to contact healthcare providers. "In medicine, we say that you have to give the right treatment to the right person at the right time," Jensen stated. By extending the therapeutic window for medical intervention, he believes this technology could save lives and significantly reduce healthcare costs by preventing hospitalizations and facilitating home management of chronic conditions.