LONDON: New strategy by Warwick University to model the spread of Covid-19 combines smartphone data captured in people’s movements and shows promises to help develop optimal locking policies.
Ritabrata Dutta from the University of Warwick, England, and colleagues present this finding in the Journal of PLO computing Biology open access.
Evidence shows that effective locking in mitigating the spread of Covid-19. However, they come with high economic costs, and in practice, not everyone follows government guidance on locking.
Thus, the Ritabrata Dutta and his colleagues propose, the optimal locking strategy will balance between controlling the ongoing Covid-19 pandemic and minimizing locking economic costs.
To help guide such strategies, the researchers developed a new mathematical model that simulated the spread of Covid-19.
Models focus on England and France and – using statistical approaches known as the estimated calculation of Bayesian – they combine public health data and data on changes to community movements, such as those captured by Google through Android devices; This mobility data serves as a measure of the effectiveness of locking policies.
Then, the researchers showed how their models could be applied to design optimal locking strategies for the UK and France using mathematical techniques called optimal control.
They show that it is possible to design an effective locking protocol that allows the reopening part of the workplace and school when taking into account the cost of public health and economic costs. Models can be updated in real-time, and they can be adjusted to any country that can be expected by public health and Google Mobility data.
“Our work opens the door to greater integration between the epidemiology model and real world data to, through the use of supercomputers, determine the best public policy to reduce the pandemic effect,” said Ritabrata Dutta.
Ritabrata Dutta added, “In the future not too far, policy makers may be able to reveal certain priority criteria, and computing machines, with different dataset use, can determine the best action.”
Furthermore, researchers plan to improve the models of their countries to work on a smaller scale; In particular, each of the 348 regional authorities in the UK.
The researchers added, “Large data integration, epidemiological and supercomputer models can help us design optimal locking strategies in real-time while balancing public and economic health costs.”
