A Bottom-up Approach to Epidemic Modeling
- Xun Shi will showcase the findings or outcomes of their project thus far as a Geospatial Fellow for advancing COVID-19 research and education.
- Sara L. McLafferty will serve as a discussant for the webinar
Xun Shi
Dartmouth College
Sara L. McLafferty
University of Illinois Urbana-Champaign
Xun Shi Dartmouth College
A Bottom-up Approach to Epidemic Modeling: Conceptual Framework, Implementation, and Case Study
Most current modeling works of COVID-19 are based on the classic SIR model, typically at the population level and adopting top-down strategies. They are sensitive to modeling setting and parameter values that feature subjectivity. Their aggregate data also mask variation across people and space. An alternative approach is to model at the individual level and to resort to a bottom-up strategy, which takes advantage of increasingly available big data of individuals’ mobilities and high-performance computing capabilities. It anticipates general and high-resolution spatiotemporal patterns of the epidemic to emerge from the modeling process. We have been developing a bottom-up approach to epidemic modeling for years. The components of the approach include: 1) disaggregating available human mobility data to obtain individual-level trajectory data, using a Markov Chain Monte Carlo (MCMC) process; 2) tracing contacts among individuals based on their daily activity trajectories; 3) building an Epidemic Forest model that represents transmission relationships among actual disease cases; and 4) predicting future epidemics based on the epidemic characteristics derived from the Epidemic Forest model, as well as data depicting intervention scenarios through agent-based modeling. The process has been preliminarily implemented within an ArcGIS environment and is being migrated to the CyberGISX platform. It has been applied to case studies of COVID-19 in China.
Xun Shi is a Professor of Geography at Dartmouth College. He has been highly active in the area of health-related geospatial research. His research covers disease mapping, disease-environment association detection, communicable disease modeling, healthcare accessibility assessment, and data infrastructure for health-GIS. He received funding support from NIH, NSF, CDC, and other sources. He published more than 70 research papers in international journals, and developed ArcHealth, a software package serving spatial analytical functions particularly requested by health-related studies and practices. He served as the Chair of the Health and Medical Geography Specialty Group of the American Association of Geographers (AAG) during 2015-2016, and as an editor of AAG Annals during 2008-2018.
Sara L. McLafferty University of Illinois Urbana-Champaign
Sara L. McLafferty is a Professor at University of Illinois Urbana-Champaign. Her current research investigates place-based inequalities in health and well-being and access to health services for women, immigrants and racial/ethnic minorities in the United States. Her ongoing work examines the impacts of increasing economic inequality and residential segregation on women's commuting times and modes and on maternal and infant health outcomes. She also uses and develops GIS and spatial analysis methods for examining health and social issues in cities and planning public health interventions. She is currently an Associate Editor of Health and Place and serves on the editorial boards of Geographical Analysis, and Spatial and Spatiotemporal Epidemiology.
A Bottom-up Approach to Epidemic Modeling
Description
Date: Mon, April 26, 2021
Time: 4:00 - 5:00 pm U.S. Central Time
Status: Event Ended
To Register for this Webinar (FREE):
- First "Log into Your Account" (see link above) using AAG credentials
- You can create a free username and password with AAG
- If you are a member with AAG you can use your existing login credentials
- If you participated in AAG Annual Meetings (or other events) you may still have active login credentials
- Once logged in successfully, click on "Register" below, to submit a registration form.
- Register
Register (FREE) to attend this webinar:Register