Quantifying the effect of quarantine control in Covid-19 infectious spread using machine learning

Raj Dandekar; George Barbastathis

Existing models analyzing the role of travel restrictions in the spread of Covid-19 either used parameters based on prior knowledge of SARS/MERS coronavirus epidemiology and not derived independently from the Covid-19 data (Chinazzi et al. 2020 ), or were not implemented on a global scale (Kraemer et al. 2020) . In this paper, we propose augmenting a first principles-derived epidemiological model with a data-driven module, implemented as a neural network. We leverage this model to analyze and compare the role of quarantine control policies employed in Wuhan, Italy, South Korea and USA, in controlling the virus effective reproduction number R t Read et al. 2020; Tang et al. 2020; Li et al. 2020a; Wu & Leung 2020; Kucharski et al. 2020; Ferguson et al. 2020) . In the original model, known as SEIR (Fang et al. 2006; Saito et al. 2013; Smirnova et al. 2019) , the population is divided into the susceptible S, exposed E, infected I and recovered R groups, and their relative growths and competition are represented as a set of coupled ordinary differential equations. The simpler SIR model does not account for the exposed population E. These models cannot capture the largescale effects of more granular interactions, such as the population's response to social distancing and quarantine policies. This is where data come in: in our approach, a neural network added as a non linear function approximator (Rackauckas et al. 2020) informs the infected variable I in the SIR model. This neural network encodes information about the quarantine strength function in the locale where the model is implemented. The neural network is trained from publicly available infection and population data for Covid-19 for a specific region under study; details are in the Materials and Methods section. Thus, our proposed model is globally applicable and interpretable with parameters learnt from the current Covid-19 data, and does not rely upon data from previous epidemics like SARS/MERS.