Modeling the Spread of Infectious Diseases on State-Level Digital Twins
Joy Kitson, University of Maryland, College Park
Preventing the spread of infectious diseases requires implementing interventions at different levels of government and evaluating the potential impact and efficacy of those pre-emptive measures. Modeling epidemic diffusion and possible interventions using agent-based modeling can further this goal. We present refinements to Loimos, a highly parallel simulation of epidemic diffusion, designed to run on realistic ”digital twin” populations. We explain how these populations are generated from a variety of data sources including US Census Public Use Microdata Sample (PUMS), National Household Travel Survey (NHTS), and American Community Survey (ACS) commute flow data. We then outline the process used to validate the simulation on a range of these digital twin populations. We additionally describe a generic framework for specifying public health interventions along with its implementation in Loimos. We demonstrate that our implementation of Loimos is able to scale to large core counts on Rivanna at the University of Virginia In particular, Loimos is able to simulate a realistic interaction network for the state of Pennsylvania in an average of 0.94 seconds per simulation day when executed on 48 nodes on Rivanna, processing around 93 billion interactions (person-person edges) in under three minutes for an average of 517million traversed edges per second (TEPS).
Authors: Joy Kitson1, Ian Costello1, Jiangzhuo Chen3, Sameer Mustaqali1, Diego Jimenez2, Stefan Hoops2, Henning Mortveit2, Esteban Meneses2, Jaemin Chin5, Jae-Seung Yeom6, Madhav V. Marathe2,4, Abhinav Bhatele1
1Department of Computer Science, University of Maryland, College Park, USA
2National High Technology Center, Costa Rica
3Biocomplexity Institute and Initiative, University of Virginia, USA
4Department of Computer Science, University of Virginia, USA
5Department of Computer Science, University of Illinois at Urbana-Champaign, USA
6Lawrence Livermore National Laboratory, USA
Abstract Author(s): (see above entries)