Nevertheless,
Mcgill team develops ai detect new:
Montreal-An artificial intelligence platform developed by researchers from McGill University is able to predict acute systemic inflammation even before the appearance of the first symptoms. Nevertheless, from data provided by different ready-to-wear technologies.
This could one day allow doctors to tackle the problem a few days earlier. However, especially in patients whose state of health is already fragile and in whom a new infection could have serious consequences.
This technology could also possibly reduce costs for the health system, avoiding complications and hospitalizations.
“We were very interested to see if the physiological data measured using laptops (…) could be used to form an artificial intelligence system capable of detecting an infection. In addition, a disease resulting from inflammation,” explained the main author of the study, Professor Dennis Jensen of the Department of Kinesiology and Physical Education of the McGill University, who discussed mcgill team develops ai detect new his premises Canadian.
“We wondered if we could detect hasty changes in physiology. Moreover, and from there predict that someone is about to get sick.”
The artificial intelligence model developed by Professor Jensen. Moreover, his colleagues uses the biometric data generated by an intelligent ring, an intelligent watch or an intelligent clothing to predict precisely an acute systemic inflammation – an early immune response to viral infections of the respiratory tract.
Even if it is a natural defense mechanism of the body which is usually resolved itself. Therefore, this inflammation can cause serious health problems, especially in populations who have pre-existing health problems.
“It is like an iceberg. For example, illustrated Professor Jensen, who ensures that this model is the only one in the world to use such physiological measures, not symptoms, to detect a problem. Nevertheless, Once the ice cracks on the surface. the symptoms are started and it mcgill team develops ai detect new is a little too late to start treating. ”
Experience
McGill researchers administered a influenza vaccine attenuated to 55 healthy adults to simulate an infection in them. The subjects were followed for a period of seven days before inoculation five days later.
Participants also carried. simultaneously and for the entire duration of the study, a ring, a watch and a garment connected to continuously monitor several physiological parameters and activities, including heart rate, variability of heart rate, body temperature, respiratory frequency, arterial pressure, physical activity and sleep quality.
The researchers also measured biomarkers of systemic inflammation using repeated blood tests. carried out PCR tests to detect the presence of respiratory pathogens and used a mobile application to collect the symptoms reported by the participants, it was explained by press release.
In total, more than two billion data has been collected to cause automatic learning algorithms. Ten different AI models have mcgill team develops ai detect new been developed. but the researchers ultimately decided not to retain for the rest of the project only the model that used the least data.
This model has properly detected almost 90 % of real positive cases and was considered more practical for daily surveillance.
Individually. Professor Jensen said, none of the physiological or activity measures from the ring, the watch or the T-shirt is sensitive enough to detect the way the body reacts.
“An increase in heart rate alone can only correspond to two beats per minute. it is not really relevant on the clinical level,” he said. The decrease in the variability of the heart rate can be very modest. The increase in temperature can be very modest. Additionally, The idea was therefore that by examining multimodal measures. several different measures, we would be able to identify subtle changes in physiology. ”
Remarkable, algorithms have also successfully detected mcgill team develops ai detect new systemic inflammation in four participants infected with SARS-COV-2 during the study. In each case. the algorithms reported the immune response up to 72 hours before the appearance of symptoms or the confirmation of the infection by PCR test.
Ultimately. researchers hope to be able to develop a system that will inform the patient of a possible inflammation so that he can then communicate with his health care provider.
“In medicine it is said that it is necessary to provide the right treatment to the right person at the right time. ” said Professor Jensen.
The other watchword in medicine is “hasty screening”: whether it is a simple cold. cancer, the more we tackle the problem early, the better the prognosis, “because once the symptoms have appeared, it begins to be made late,” he recalled.
By widening the therapeutic window inside which we can intervene. he added, one could save lives mcgill team develops ai detect new and make great savings by avoiding hospitalizations and allowing the management at home of chronic problems or even aging.
“In a way, we hope to revolutionize personalized medicine,” concluded Professor Jensen.
The conclusions of this study were published by the newspaper The Lancet Digital Health.
Mcgill team develops ai detect new
Further reading: The microbiome or the challenges of a new era of personalized nutrition for health – A Tiktok trend puts the health of young people at risk – “Silent” and “Living the day”, the tiger mosquito extends its territory to Deux-Sèvres – Can intermittent fasting prevent cognitive decline? – The new sun protection from Korea which completes your SPF cream.
