A data-driven network model for the emerging COVID-19 epidemics in Wuhan, Toronto and Italy

Abstract

The ongoing Coronavirus Disease 2019 (COVID-19) pandemic threatens the health of humans and causes great economic losses. Predictive modeling and forecasting the epidemic trends are essential for developing countermeasures to mitigate this pandemic. We develop a network model, where each node represents an individual and the edges represent contacts between individuals where the infection can spread. The individuals are classified based on the number of contacts they have each day (their node degrees) and their infection status. The transmission network model was respectively fitted to the reported data for the COVID-19 epidemic in Wuhan (China), Toronto (Canada), and the Italian Republic using a Markov Chain Monte Carlo (MCMC) optimization algorithm. Our model fits all three regions well with narrow confidence intervals and could be adapted to simulate other megacities or regions. The model projections on the role of containment strategies can help inform public health authorities to plan control measures.

Keywords: COVID-19; Control measures; Heterogeneity; Mitigation strategies; Network model.

Fig. 1

Fitting the number of reported new cases and the cumulative number of reported cases between Jan 11, 2020 and Mar 31, 2020 for Wuhan on Watts–Strogatz network.(A) Fitting the number of reported new cases on the Watts–Strogatz network. (B) Fitting the cumulative number of reported cases on the Watts–Strogatz network.

Fig. 2

Fitting the number of reported new cases and the cumulative number of reported cases for Toronto on the Watts–Strogatz network. (A) Fitting the number of reported new cases on the Watts–Strogatz network. (B) Fitting the cumulative number of reported cases on the Watts–Strogatz network.

Fig. 3

Fitting the number of reported new cases and the cumulative number of reported cases for Italy on the Watts–Strogatz network. (A) Fitting the number of reported new cases on the Watts–Strogatz network. (B) Fitting the cumulative number of reported cases on the Watts–Strogatz network.

Fig. 4

The impact of the variability on the healthcare capacity on the spread of the epidemic in Wuhan on the Watts–Strogatz network.

Fig. 5

The impact of mitigation strategies on the spread of COVID-19 epidemic in Toronto on the Watts–Strogatz network. In this figure and the following figure, the dashed lines represent $95\text{\%}$ confidence intervals. In (A) and (B), the red, purple, and green lines represent that the transmission rates are unchanged, reduced by $20\text{\%}$, and reduced by $40\text{\%}$, respectively. In (C) and (D), the red, purple, and green lines represent the rate of quarantine, $q=0$, 1/8, and 1/4, respectively. In (E) and (F), the red, purple, and green, and light blue lines represent that the node degrees of symptomatically infected individuals are reduced by $0$, $1$, $2$, and $3$, respectively. (A) The number of newly infected individuals after reducing the transmission rates by personal protection and social distancing. (B) The cumulative number of infected individuals after reducing the transmission rates by personal protection and social distancing. (C) The number of newly infected individuals after close contacts are quarantined. (D) The cumulative number of infected individuals after close contacts are quarantined. (E) The number of newly infected individuals after the node degrees of symptomatically infected individuals are reduced. (F) The cumulative number of infected individuals after the node degrees of symptomatically infected individuals are reduced.

Fig. 6

The impact of mitigation strategies on the spread of COVID-19 epidemic in Italy on the Watts–Strogatz network. (A) The number of newly infected individuals after reducing the transmission rates by personal protection and social distancing. (B) The cumulative number of infected individuals after reducing the transmission rates by personal protection and social distancing. (C) The number of newly infected individuals after close contacts are quarantined. (D) The cumulative number of infected individuals after close contacts are quarantined. (E) The number of newly infected individuals after the node degrees of symptomatically infected individuals are reduced. (F) The cumulative number of infected individuals after the node degrees of symptomatically infected individuals are reduced.