Jinxing Guan

Modelling and predicting the behaviour of infectious diseases is essential for early warning and evaluating the most effective interventions to prevent significant harm. Compartmental models produce a system of ordinary differential equations (ODEs) that are renowned for simulating the transmission dynamics of infectious diseases. However, the parameters in compartmental models are often unknown, and they can even change over time in the real world, making them difficult to determine. This study proposes an advanced artificial intelligence approach based on physics-informed neural networks (PINNs) to estimate time-varying parameters from given data for the compartmental model. Our proposed PINNs method captures the complex dynamics of COVID-19 by integrating a modified Susceptible-Exposed-Infectious-Recovered-Death (SEIRD) compartmental model with deep neural networks. Specifically, we modelled the system of ODEs as one network and the time-varying parameters as another network to address significant unknown parameters and limited data. Such structure of the PINNs method is in line with the prior epidemiological correlations and comprises the mismatch between available data and network output and the residual of ODEs. The experimental findings on real-world reported data data have demonstrated that our method robustly and accurately learns the dynamics and forecasts future states. Moreover, as more data becomes available, our proposed PINNs method can be successfully extended to other regions and infectious diseases.

Background: In late February 2022, the Omicron epidemic swept through Shanghai, and the Shanghai government responded to it by adhering to a dynamic zero-COVID strategy. In this study, we conducted a retrospective analysis of the Omicron epidemic in Shanghai to explore the timing and performance of control measures based on the eventual size and duration of the outbreak. Methods: We constructed an age-structured and vaccination-stratified SEPASHRD model by considering populations that had been detected or controlled before symptom onset. In addition, we retrospectively modeled the epidemic in Shanghai from 26 February 2022 to 31 May 2022 across four periods defined by events and interventions, on the basis of officially reported confirmed (58,084) and asymptomatic (591,346) cases. Results: According to our model fitting, there were about 785,123 positive infections, of which about 57,585 positive infections were symptomatic infections. Our counterfactual assessment found that precise control by grid management was not so effective and that citywide static management was still needed. Universal and enforced control by citywide static management contained 87.65% and 96.29% of transmission opportunities, respectively. The number of daily new and cumulative infections could be significantly reduced if we implemented static management in advance. Moreover, if static management was implemented in the first 14 days of the epidemic, the number of daily new infections would be less than 10. Conclusions: The above research suggests that dynamic zeroing can only be achieved when strict prevention and control measures are implemented as early as possible. In addition, a lot of preparation is still needed if China wants to change its strategy in the future.

Since late 2019, the beginning of coronavirus disease 2019 (COVID-19) pandemic, transmission dynamics models have achieved great development and were widely used in predicting and policy making. Here, we provided an introduction to the history of disease transmission, summarized transmission dynamics models into three main types: compartment extension, parameter extension and population-stratified extension models, highlight the key contribution of transmission dynamics models in COVID-19 pandemic: estimating epidemiological parameters, predicting the future trend, evaluating the effectiveness of control measures and exploring different possibilities/scenarios. Finally, we pointed out the limitations and challenges lie ahead of transmission dynamics models.