Project Summary
The recent SARS-CoV-2 pandemic has highlighted that mathematical modeling of infectious disease is critical
for data-informed decision making. At the same time, however, it has been made clear that the modeling
community does not have appropriately advanced informatics infrastructures that facilitate a rapid consensus
understanding during epidemics and that put the power of modeling in the hands of local public health
stakeholders. This project proposes three integrated elements to transform the workflow of constructing, testing,
and crowd-sourcing spatial epidemiological models to gain deep understanding of epidemics, to provide usable
decision-making tools for local stakeholders, and to propose concrete, locally focused solutions. Our proposal is
to develop a proof-of-concept, collaborative informatics framework for model construction, analysis and
comparison, followed by rigorous optimization of spatial intervention strategies. In Aim 1, we design EpiMoRPH
(Epidemiological Modeling Resources for Public Health), a system that will streamline and automate the
construction and testing of spatial models against benchmark data. EpiMoRPH will support rapid model
comparisons in a community-driven environment to build consensus and to produce a broad understanding of
which modeling approaches are most appropriate in different spatial contexts. Importantly, EpiMoRPH will assist
local public health stakeholders with deciding on the best, community-contributed models that are relevant for
their particular situations and will then implement those best models to make locally customized forecasts. In
Aim 2, we make advances in the automation of spatial and robust optimization algorithms, with the goal of
allowing non-expert users to generate tailor-made intervention strategies relevant to their local municipalities.
Here, we will develop a tool kit of robust optimization algorithms that account for various uncertainties and that
will gradually build upon the functionality of EpiMoRPH. Importantly, a driving motivation for this tool kit is to
ensure that the optimization routines allow public health stakeholders to balance the control of transmission and
disease outcomes with the equitable allocation of interventions across racial, ethnic, and socio-economic
sectors. In Aim 3, we will collaborate with a Public Health Advisory Council to test, formally evaluate, and refine
our model-based technologies, ensuring that our innovations meet the needs of public health partners, while
also appealing to the broader community of epidemiological modelers. Together our aims will build accessible
and sustainable technologies that put epidemiological modeling and optimization methods in the hands of local
public health decision-makers.
