Computational models and data- and model-driven computer simulations for disease spreading are increasingly critical in predicting geo-temporal evolution of epidemics and managing health emergencies. These models and simulations, however, require access to, integration, and analysis of large volumes of data, including demographic, mobility, and health-care data and models, from diverse sources. Moreover, because of the volume and complexity of the data, the varying spatial and temporal scales at which the key processes operate and relevant observations are made, experts lack the means to adequately simulate these processes in real-time and assess the robustness of conclusions driven from the resulting simulations. While very powerful and highly modular and flexible epidemic spread simulation software exists, these suffer from two major challenges that prevent wide-spread usage and real-time decision making and, consequently, reduce potential for transformative impact: cost of modeling and cost of execution. Therefore, our goal is to address the key data challenges underlying epidemic spread simulations, which, today, hinder effective scenario analysis and real-time decision making. Intellectual Merit: In developing the proposed epidemic simulation data management system (epiDMS), we will tackle two key challenges that render data-driven disease spreading simulations, today, difficult: The key characteristics of many data sets of urgent interest to data-intensive simulations include the following: (a) voluminous, (b) multi-variate, (c) multi-resolution, (d) multi-layer, and (d) geo-temporal. We argue that significant savings can be obtained by supporting modular re-use of existing simulation results in new settings. This requires (a) segmentation and indexing of real-time disease data and simulation along with the corresponding building models and other contextual metadata and (b) re-contextualization and modular re-composition or sketching of disease spread models and new simulation traces based on new building geo-temporal and contextual metadata. Therefore, this proposal will impact computational challenges that arise from the need to model, analyze, index, visualize, search, and recompose, in a scalable manner, large volumes of multi-variate and multi-layer time series underlying epidemic spread simulations. The proposed multi-resolution data encoding, partitioning, and processing algorithms are efficiently computable, leverage massive parallelism, and result in highly compact data descriptions. Broader Impacts: Recent global pandemics, which had immense economic impact, highlighted the importance of real-time analysis of the spatio-temporal dynamics of emerging infectious diseases. The proposed epiDMS will fill an important hole in data-driven decision making during health-care emergencies and, thus, will enable applications and services with significant economic and health impact. Chowel (co-PI) is an associate professor in the School of Human Evolution and Social Change (SHESC) , with affiliations in Thematical and Computational Modeling Sciences Center, the Center for Global Health, and the Mate Center for Population Dynamics. In addition to being a Professor of Computer Science and Eng., Candan (PI) is also a Senior Sustainability Scientist at ASUs Global Institute of Sustainability, one of whose missions is to connect researchers with businesses, industry, municipalities, and government. We will complement our collaboration by leveraging the Institutes connections for dissemination and capacity building. The educational impact of the proposed project will be on graduate education through mentoring of PhD students and on CSE and SHESC curricula through incorporation of research challenges and outcomes into existing undergraduate and graduate classes. This will introduce computer science students to geo-spatial and temporal data management, indexing, and analysis, as well as will familiarize them with challenges underlying managing health emergencies. ASU recruits top-quality undergraduates through a nationally recognized residential Honors College and the Minority Access to Research Careers program and we will recruit undergraduate research interns in the project, with the aim of inspiring them towards graduate research. epiDMS will also be a testbed for undergraduate Capstone Projects. Since ASU has one of the highest Hispanic student populations in the nation, we will seek to recruit CSE Hispanic students to participate in the project. The PIs have a track record in advising female students and postdocs (including 4 PhD and 1 MS students currently) and will continue to recruit female and minority students. Also, through our collaborations with the ASUs Disability Resources Administration, we will recruit students who are visually impaired or otherwise disabled. Keywords: Epidemic spreading, Simulation data sets, Data and model repurposing, Health emergencies
|Effective start/end date||9/1/13 → 8/31/18|
- NSF: Directorate for Biological Sciences (BIO): $515,602.00
Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.