US researchers have developed a portable monitoring device capable of detecting coughs and the size of a crowd to follow the evolution of the spread of a respiratory virus in real time.
As the whole world is paralyzed by the Covid-19 pandemic, the news is timely. American researchers have announced that they have made a portable monitoring device capable of detecting the cough and the size of a crowd to follow in real time the evolution of the spread of a virus, announced the University of Massachusetts (United States) on March 19 in a press release. This device, FluSense, will be used in hospitals, health service waiting rooms and large public spaces. Ultimately, it could help predict seasonal flu and other viral respiratory epidemics like the one we are currently facing.
“I have been interested in non-vocal bodily sounds for a long time. I thought if we could capture the sounds of coughing or sneezing in public spaces where lots of people naturally congregate, we could use that information as a new source of data to predict epidemiological trends.”explains Tauhidur Rahman, assistant professor of computer science and information sciences, co-author of the study, in the preamble.
So he and his colleagues started by developing a cough model in the lab. Then they trained a computational model to draw boundaries on thermal images of people and count them. “Our main objective was to build predictive models at the population level, not at the individual level”details Tauhidur Rahman, assuring that this platform does not store any personally identifiable information.
Accurately predict daily disease rates
The scientists then placed the devices, locked in a rectangular box, in four waiting rooms of the health services of the University of Massachusetts in Amherst (United States). From December 2018 to July 2019, the platform collected and analyzed over 350,000 thermal images and 21 million non-vocal audio samples from public waiting rooms.
Result: FluSense was able to accurately predict the clinic’s daily illness rates. “The first symptom-related information captured by FluSense could provide valuable additional and complementary information to current flu forecasting efforts.”, such as the FluSight Network, a multidisciplinary consortium of influenza forecasting teams, the researchers note. “This can allow us to predict flu trends much more accurately.”welcomes Rahman.
For Forsad Al Hossain, lead author of the study, FluSense is a good illustration of the possibilities of artificial intelligence in the field of health. “We try to bring machine learning systems to the forefronthe explains, showing the press the compact components inside the FluSense device. All the processing is done here. These systems are becoming cheaper and more powerful.”
Determine vaccination campaign schedules
Going forward, FluSense should be tested in other public areas and geographic locations. “We have initial validation that cough does indeed correlate with influenza-related illnesses (…) Now we want to validate it beyond this specific hospital setting and show that we can generalize to other places”says vector-borne disease expert Andrew Lover.
A little over a month ago, a Canadian company by the name of BlueDot had already been talked about for somewhat similar reasons. Using data available online and machine learning systems, she had detected the first signs of a coronavirus infection in Wuhan, in southern China, from December 31, well before the WHO alerted the whole world to the subject. She then predicted that the infectious agent would pass the following days from the cradle of contamination in Bangkok, Seoul, Taipei and Tokyo.
Eventually, models of this kind could help health authorities in different countries determine the timing of vaccination campaigns, possible travel restrictions or the allocation of medical supplies.
.