epiDMS: Data Management and Analytics for Decision-Making From Epidemic Spread Simulation Ensembles

Abstract

Background: Carefully calibrated large-scale computational models of epidemic spread represent a powerful tool to support the decision-making process during epidemic emergencies. Epidemic models are being increasingly used for generating forecasts of the spatial-temporal progression of epidemics at different spatial scales and for assessing the likely impact of different intervention strategies. However, the management and analysis of simulation ensembles stemming from large-scale computational models pose challenges, particularly when dealing with multiple interdependent parameters, spanning multiple layers and geospatial frames, affected by complex dynamic processes operating at different resolutions.

Methods: We describe and illustrate with examples a novel epidemic simulation data management system, epiDMS, that was developed to address the challenges that arise from the need to generate, search, visualize, and analyze, in a scalable manner, large volumes of epidemic simulation ensembles and observations during the progression of an epidemic.

Results and conclusions: epiDMS is a publicly available system that facilitates management and analysis of large epidemic simulation ensembles. epiDMS aims to fill an important hole in decision-making during healthcare emergencies by enabling critical services with significant economic and health impact.

Keywords: analytics; big data; data management; epidemics; public health decision-making; simulation ensembles.

Figure 1.

epiDMS overview.

Figure 2.

A sample epiDMS screenshot, which includes scenario-based querying and exploration. The figure shows a query posed to epiDMS, the set of results (visualized in the form of a navigable hierarchy of heat maps), and 2 simulations selected for detailed comparison. Please see the accompanying Supplementary Materials and the video available at https://www.youtube.com/watch?v=9w-4nDhXv3k for more details.