Efficient Strategies for Pandemic Monitoring and Recovery

In the absence of effective vaccines or treatments, social distancing has been shown to be effective in controlling the initial spread of a pandemic. While the decision on when to start social distancing can be based on the occurrence of a few positive test cases, the decision on when to ease such measures during pandemic recovery is considerably more difficult due to the possibility of a second wave. Our goal in this project is to develop new artificial intelligence (AI) algorithms to support this decision-making, by making more efficient use of limited test availability and also drawing on novel data sources (and physical models) not limited by testing. The research will be divided into two inter-related thrusts. The first thrust in on efficient ways to determine the status of individuals in the community through new AI based methods for group testing. The second thrust is focused on community level pandemic assessment using a data-driven quickest change detection (QCD) framework.

The project, if successful, will have a direct and obvious impact on pandemic recovery, for COVID-19 and future pandemics. This will in turn have significant social and economic impacts. The goal is to rapidly deploy research results via relationships in government and industry. Broader impact activities include: navigating the ethical tradeoff between equity and accuracy in group testing designs; supporting a diverse cohort of undergraduate researchers in this topical area; developing a new, general educational module for laboratory-based data science classes (at Illinois and around the world); integrating diversity by training a diverse cohort of graduate students; and public outreach via established media presence.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.