We conducted work on this project by carrying out all five work packages (WPs). In WP1, we developed a stochastic model to study the epidemic spreading problem, addressing the impact of pre-symptomatic transmission on heterogeneous contact network structures which are relevant in COVID-19. We applied the advanced dynamic message-passing method for solving this model, which provides accurate and efficient algorithms. We also derived analytical expressions for epidemic thresholds and discussed their implications for different containment strategies, e.g. isolating individuals and reducing transmission through social distancing and face masks. In WP2, we proposed and validated group testing strategies to boost the testing capacity through collaboration with researchers in the SARS-COV-2 PCR testing labs at the University of Birmingham and colleagues at King’s College London, and provided a method to reduce the false negative rate of two-stage group testing. In WP3, we studied the mathematical model of failure spreading of support service networks due to epidemic spreading such as COVID-19. We developed algorithms to solve this model using dynamic message-passing and devised mitigation strategies using the optimal control framework. In WP4, we investigated decision-making strategies under uncertainty that is relevant at the early stage of a pandemic such as COVID-19. In WP5, we applied the developed mathematical techniques in neighboring fields in complex systems such as routing and machine learning. For training and knowledge transfer, the Fellow attended five training workshops organized by the Research and Knowledge Exchange Team of the host university. He actively engaged in various teaching and supervision activities, including delivering tutorials, marking assignments and supervising MSc students on research problems. The Fellow participated in grant management, proposal preparations and risk management through the guidance of the host supervisors. The Fellow provided referee services for scientific journals, including Journal of Statistical Mechanics: Theory and Experiment, Scientific Reports and IEEE Access.
The academic results in this project were mainly disseminated through publications in scientific journals, which include (1) one published paper about epidemic-spreading with pre-symptomatic transmission; (2) one forthcoming paper about failure spreading of support service networks due to epidemic spreading; (3) one forthcoming paper about decision-making strategies under uncertainty; (4) three published papers and one forthcoming paper in other complex systems. Before the COVID-19 pandemic, we were invited to three conferences to present our work. After the onset of the pandemic, we attended and presented in online conferences and used other online platforms (such as researchgate) to communicate our work to other researchers. We also tried to lobby the government of the United Kingdom to adopt group testing strategy to boost test capacity during the first wave of the COVID-19 pandemic.
