The United States predicts high-risk sites for the spread of the new coronavirus

BEIJING, Nov. 11 (Xinhua Zhang Mengran) -- Reopening restaurants, gyms, cafes and hotels poses the greatest risk of new coronavirus transmission, according to an epidemiological modeling study based on big data from the United States published in the British journal Nature on Monday. Models show that reducing the use of these venues could significantly reduce the number of people infected.
To assess how changes in human mobility could change the spread of the new coronavirus, the Stanford University team used mobile phone data from the United States (collected from March 1 to May 2, 2020) to map the movements of millions of people in different communities. They combined this data with a new model of coronavirus transmission to identify potentially high-risk sites and at-risk populations. Their model accurately predicts the number of confirmed cases per day in the 10 largest metropolitan areas, such as Chicago, New York City and San Francisco.
the fineness of the moving data allowed researchers to simulate hourly infections in nearly 553,000 different locations in 20 categories that people often visit, also known as "points of interest." The model predicts that a small number of these locations, such as full-service restaurants, contribute the majority of infections.
, for example, in the Chicago metropolitan area, 10 percent of "points of interest" contributed 85 percent of all "points of interest" to predict infections. The model also predicts that low-income groups are more susceptible to infection than high-income groups because they are unable to move significantly and they go to smaller and more crowded places, which also increases the risk of infection. For example, low-income groups regularly visit convenience stores with 59 percent more people per square foot than high-income groups, and customers stay 17 percent longer on average.
by simulating who is susceptible to infection in which and where, the researchers also assessed the effectiveness of different reopening strategies. In their view, this model could serve as a reference for the development of a re-opening policy.
, for example, model projections show that keeping site usage at 20 percent of its maximum capacity can reduce new infections by more than 80 percent, but the total number of customers arriving will only decrease by 42 percent.