Ljubljana – A team of researchers led by Jure Leskovec, the Slovenian Stanford computer scientist, has created a computer model to predict the spread of Covid-19 based on people’s travel habits and movements. They published their article in the journal Nature.
The Covid-19 pandemic has drastically changed the way people behave and their mobility patterns, however, epidemiologic models that would accurately predict the spread of the novel coronavirus based on people’s mobility patterns and lockdown restrictions do not exist yet.
This is why the team developed an epidemiological model to model the spread of coronavirus based on travel habits and people’s movements.
They created the model based on data on mobile phone locations capturing the movement of almost 100 million people in 10 major US cities.
They accurately predicted the spread of Covid-19 this spring by analysing three factors that drive infection risk: where people go in the course of a day, how long they linger and how many other people are visiting the same place at the same time.
Their approach can assess every hour of the day the mobility network, how many people from each neighbourhood visit various locations such as restaurants, stores, gyms, churches or schools and for how long.
They found that without any of the coronavirus measures a third of the US population would be infected in a month, while they also established the existence of “superspreader” sites, that is about 10% of locations, where more than 85% of all virus transmissions occur.
Through their approach they also explained why the socially deprived people are more likely to catch the virus, because they tended to stay at home less and they moved at more diverse locations.
The researchers say their model could serve as a tool for officials as they take decisions to lift restrictions.