Researcher New York University Before the proliferation of COVID-19 was realized, we devised a new way to help predict it. This is an approach that can cost-effectively control the spread of the virus until the vaccination program is complete.
By analyzing online searches for mobile and isolated activities, their preliminary AI tools can predict where and when outbreaks may occur. Authorities with this information could take emergency actions, such as those recently seen in Victoria, to sprouting the virus and reduce its impact on local businesses before it is replicated to uncontained levels. I can do it.
Professor Megan Coffee and Professor Anasse Bari of New York University say the technology is based on “alternative data.” This is a concept previously used in finance to generate data-driven investments, such as predicting business revenue from satellite images of parked cars.
“What we wanted to do here was to develop a barometer of behavior,” said Professor Coffee. Hospital + healthcare.. “When modeling an outbreak of an infectious disease, the curves are always nicely smooth, but in the epidemic reality I’ve worked on, there are ups and downs, and in some cases even roller coasters soaring and plunging. These changes are many. Human behavior is a significant cause, although it may be due to factors. “
“As COVID-19 becomes more widespread, if someone googles for bar closing times or searches for directions to the gym, they give insight into what risks they may take in the future. “Masu,” added Professor Bali.
“At population size, these searches can be used to create data analysis tools that can be a barometer of behavior. Extend data analysis tools to learn from ensemble of data sources and increase transmissions such as open tables. You can track indoor activity to track restaurant volume, bike and taxi data, and real estate data to measure mobility and mixing. “
The tools are still in the preparatory stage and vary by geographic region, age, background, and epidemic stage and need to be tweaked to better explain the preferences of different activities.
However, this is not the first time researchers have successfully used AI to improve their nationwide pandemic response. Last year, they focused on tools that could predict which COVID-19 patients would develop a serious illness. This is a solution that helps clinicians determine their acute care priorities.
Next, researchers are working to extend the technology to other data sources that can better capture more signals and predictors of a pandemic.
“This pandemic requires a holistic approach to combat it. Predict clinical severity using AI, Building early signal tools with alternative data that help predict pandemic curve spikes and falloffs, “Concluded Professor Bali.
New AI technology foretells COVID-19 outbreaks Source link New AI technology foretells COVID-19 outbreaks