Search
Friday, October 07, 2022

Your Google searches and tweets could help predict the next disease outbreak

Looks like there’s another punchline for anyone joking about the pandemic life of the past two years. But for scientists looking to predict disease outbreaks in the future, it is important data.

Scented candles started receiving an influx of negative reviews online in 2020. Dissatisfied customers declared that some of the most scented, most popular products from well-known companies like Yankee Candle had “no smell” or even smelled bad.

It wasn’t just some bad review. The most popular scented candles sold on Amazon were receiving an average of 4 to 4½ stars before 2020, but during that first year of the pandemic, rave reviews About a full star fell, Social media users wondered a link between these negative reviews and the loss of sense of smell associated with COVID-19 infection.

When COVID-19 cases rise again in late 2021 due to Omicron version, researchers note one more raise In those negative “no smell” reviews.

what are those negative online reviews Mauricio Santilana It’s called “breadcrumbs”. As people navigate the digital world, they leave traces of what is happening in their offline lives, explains the director of the Machine Intelligence Group for the Betterment of Health and the Environment (MIGHTE) at the Network Science Institute in the Northeast. Those “breadcrumbs” leave a mark for researchers like Santalana as they anticipate possible future outbreaks of COVID-19 and other diseases.

If there are discrepancies in online trends—a spike in Google searches of shops offering chicken noodle soup, a sudden influx of tweets about a family member giving up, or bad reviews on scented candles—it could indicate trouble is brewing. has been So Santilana is building machine-learning models to detect anomalies, decipher these clues, and create an early warning system for disease outbreaks.

By adding human behavior to the mix, “we’re building an observatory of disease activity using different telescopes,” says Santilana, a professor of physics and electrical and computer engineering who recently joined Northeast from Harvard University.

Your Google searches and tweets could help predict the next disease outbreak

Mauricio Santilana, director of the Machine Intelligence Group for the Betterment of Health and the Environment (MIT) at the Network Science Institute of the Northeast, and professor of physics and electrical and computer engineering. Photo by Matthew Moduno/Northeast University

Santillana is working together alessandro vespignaniDirector of the Network Science Institute and the Sternberg Family Distinguished Professor at Northeast, who lead a team of infectious-disease modelers who are developing a set of guesses About the possible future of the COVID-19 pandemic since the crisis began.

Vespignani’s model integrates details such as count of cases, hospitalizations, deaths, human mobility patterns, how often humans interact, how the virus is transmitted and more data focused on the disease. Santilana says his research adds a different kind of thermometer by looking at digital traces of human behavior that is a step removed from epidemiological data.

“In a way, we are trying to bring these two perspectives together to provide a more complete picture of outbreaks like COVID-19,” Santilana says.

Santillana and Vespignani are already collaborating, combining this digital behavioral data with epidemiological data in their modeling work. In a paper published last year in Science Advances, they showed that such a cohesive early warning system could anticipate a bounce by two to three weeks in COVID-19 cases and deaths. Joining Santalana’s Network Science Institute, the pair will work together to further develop this early-alert system for disease outbreaks – not just COVID-19.

What if you could get vaccinated against the virus that causes COVID-19 with an inhaler instead of a needle?  That's the basis behind new research by Northeast's Paul Whitford.  Photo Illustration by Alyssa Stone/Northeast University

The data that Sentilana collects includes a vast, diverse collection of information — not just Google search trends, social media posts, and online shopping reviews or orders. They have also used anonymized smart thermometer data to identify when certain types of disease can last in an area, anonymized mobility data from smartphones that show when more people may be sick, as well as for certain types. Trends in doctor searches for treatments or symptoms.

Even Google searches and social media posts cover a wide range of data. People may be searching for more information about their symptoms or quarantine recommendations, or they may simply be trying to figure out where to buy cough syrup or soup.

Increase in only one of these behaviors in an area might Indicates that COVID-19 or some other infectious disease is spreading in a community, or it could be that there was a new sci-fi movie that came out and piqued the curiosity of the people about the pandemic in general. So Santilana says it’s important to take several different data sources into account for their models. The machine learning model is also designed to detect whether the increase in certain Google searches, for example, is actually related to infections and hospitalizations to determine whether it is due to disease outbreaks. Worth considering as a precursor.

This new type of “telescope,” as Santilana called it, would be a component of America’s new disease prediction initiative, Forecast and Outbreak Analysis Center (CFA). Santalana is part of a team of experts advising that effort.

“In the same way weather forecasting systems work around the world,” he explains, “the idea is to see the information generated in real time and design systems to contribute in a variety of ways that will recognize when something unusual happens.”

Like weather forecasting agencies, CFAs will essentially be an early warning system that identifies when and where disease outbreaks may occur so that public-health officials can take action to prevent them from becoming catastrophic.

for media inquiriesplease contact shannon nargi [email protected] or 617-373-5718.

Nation World News is the fastest emerging news website covering all the latest news, world’s top stories, science news entertainment sports cricket’s latest discoveries, new technology gadgets, politics news, and more.

Latest News

Related Stories