Using a computer model to map frequent locations on a daily basis in large cities The computer model suggests that most COVID-19 infections occur at “supersprider” sites such as full-service restaurants, gyms, and cafes.
A report published in the journal Nature on Tuesday examined data from 98 million Americans collected in 10 major U.S. cities, including San Francisco, for the two months beginning in March. A team led by Stanford University then provided the information in an epidemiological model developed by Dell.
Jared Leskovac, a computer scientist at St Ledford, who led the study, told Stanford News that the model analyzed people from different demographic backgrounds and neighborhoods visiting more or less crowded sites.
“Based on all of this, we can predict the likelihood of new infections occurring anywhere or at any time,” he said.
Those predictions will later prove to be accurate based on the number of infections officially recorded by cities.
Scientists used data provided by the Denver-based company SafeGraph, which collects anonymous location information from cellphone apps one day to find out which public places people visit every day and how long they have been there. Square footage of each establishment was recorded to determine the density of hourly employment.
The various scenarios corresponding to the model, in which some industries were to be restarted but others were not, showed that the opening of full-capacity restaurants results in the greatest increase in infection. Giff was second, followed by cafes and hotels / motels. According to one scenario, if capacity was limited to 20% in all locations, new infections would be reduced by more than 80%.
When the census is combined with demographic data, the data suggests why people from poorer neighborhoods enter into the Kovid-1 contract:
Home They are less able to work from home.
The stores they shop for are more crowded than the affluent areas.
Higher They stay longer within these stores (on average about 17% longer) than in high-income areas.
These findings could help cities develop strategies for spreading COVID-19 while limiting the damage to their economies.
However, two scientists at Oxford University said more research is needed to test whether the model identifies the actual location of the infection.
Christopher Dye, an epidemiologist at the university, told Nature that the research, while promising, is “an epidemiological hypothesis” that needs to be validated with real-world data. Is required.
This study is based on previous attempts using a computer model to predict how the virus spreads in enclosed spaces. For example, the University of Colorado, Boulder, atmospheric chemist Jose Luis Jimenez has developed a tool called the Covid Airborne Transmission Estimator, which analyzes the possibility of indoor transmission of a virus through aerosols in a variety of home conditions. .
