Montreal – Researchers at the University of McGill say they have developed an artificial intelligence platform that can guess that when a person is going to come with a respiratory tract infection before feeling ill.
The study included a ring, a clock and a T-shirt participants who were equipped with sensors that recorded their biometric data. By analyzing the data, researchers were able to make accurate prediction of acute systemic inflammation -an early sign of respiratory infections such as Covid -19.
Published in Lancet Digital Health, the study states that the AI platform can one day help doctors to address health problems, which can usually solve health problems compared to earlier, especially in patients who are delicate and for whom a new infection may have serious consequences. It can also reduce costs for health care systems by preventing complications and hospitalization.
“We were very interested to see whether physical data is measured using a weedable sensor … can be used to train an artificial intelligence system that is able to detect inflamed infection or disease,” the prominent writer of the study, Professor Danis Jenson, the Professor of McGill University, McGill University’s Professor Danis Jensen.
“We were surprised whether we could detect initial changes in physiology and, from there, predict that a person is going to get sick.”
Jens says that the AI model is formed by his team, which is the first to use physical measures in the world – in which heart rate, heart rate variability, body temperature, respiratory rate, blood pressure – instead of symptoms, to detect a problem.
Acute systemic inflammation is a natural defense mechanism of the body that usually resolves on its own, but it can cause serious health problems, especially in the population with already existing conditions.
“The whole idea is like an iceberg,” Jensen said. “When the ice cracks the surface, this is when you are symptomatic, and then it is really too late to treat it.”
During the study, McGill researchers conducted a weak flu vaccine to 55 healthy adults to simulate their body infections. The subjects were monitored seven days before the vaccination and five days later.
The participants wore a smart ring, smart watch and a smart T-shirt together in the entire study. Also, researchers collected biomars in systemic inflammation using blood samples, PCR tests to detect the presence of respiratory pathogens to collect the symptoms reported by the participants.
Overall, more than two billion data points were collected to train the machine learning algorithm. Ten separate AI models were developed, but researchers chose the model that used the least data for the remaining of the project. The chosen model correctly detected about 90 percent of real positive cases and considered more practical for daily monitoring.
On his own, Jensen said, “None of the data collected from the ring, watch, or T-shirt is enough to find out how the body is reacting.”
He said, “The increase in heart rate can only correspond to two beats per minute, which is not really clinically relevant,” he explained. “Reduction in heart rate variability can be very modest. The increase in temperature can be very modest. So the idea was that by looking at many different measurements, we would be able to identify subtle changes in physiology.”
During the study, algorithms also successfully detected systemic inflammation in four participants infected with Covid-19. In each case, the algorithm flagged the immune response for 72 hours from the appearance of symptoms or confirmed the infection by PCR test.
Ultimately, researchers are expected to develop a system that will inform patients with potential inflammation so that they can contact their health care provider. “In medicine, we say that you have to give the right treatment to the right person at the right time,” Jensen said.
By expanding therapeutic window in which doctors can intervene, they said, they can save life and get important savings by enabling hospitalization and even older conditions or even aging home management.
“In a way, we hope to revolutionize personal medicine.”
This report of Canadian Press was first published on 30 July 2025.
Jean-Benoite Legolt, The Canadian Press
