Data-Driven Modeling to Improve Understanding of Human Behavior, Mobility, and Disease Spread

数据驱动建模以提高对人类行为、流动性和疾病传播的理解

• 批准号: 2109647
• 负责人: Taylor Anderson
• 金额: $ 229.38万
• 依托单位: George Mason University
• 依托单位国家: 美国
• 项目类别: Continuing Grant
• 财政年份: 2021
• 资助国家: 美国
• 起止时间: 2021-05-15 至 2026-04-30
• 项目状态: 未结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2109647&HistoricalAwards=false
• 关键词: Data; Driven; Modeling; Improve; Understanding

Models of disease dynamics are important tools used to predict the numbers of cases and deaths over time and to support policymakers as they prepare for and respond to infectious disease outbreaks. However, despite significant advances, many models still lack realistic representations of human behavior and mobility, which are key drivers of disease spread. Without accounting for the complexity of human behavior, models are limited in their ability to make accurate predictions, especially over longer time horizons. This project investigates the inclusion of realistic human behavior and mobility in models of disease spread to 1) better explain the different ways that humans respond to disease outbreaks, 2) improve predictions of infectious disease spread, and 3) help to prescribe the most effective mitigation policies. The investigators use publicly available data so that models can be rapidly deployed for any county or state in the U.S. to predict and mitigate future outbreaks of infectious respiratory diseases (e.g., COVID-19, seasonal influenza, measles, and smallpox). These models may provide more timely and accurate predictions to help the general public, key institutions, and policymakers anticipate what is to come and provide support for evidence-based policy making. This project will support professional development opportunities for early-career researchers and training opportunities for a postdoctoral researcher, graduate, undergraduate, and high school students in the Aspiring Scientists Summer Internship program.The researchers will use a data-driven approach to explain the spatio-temporal variations in the behavioral response to a disease outbreak. They hypothesize that regional variables such as average income, age, political leaning are associated with spatial patterns of behavioral response, and will leverage very large data sets to mine association rules between such variables and observed behavioral response, including social distancing, stay-at-home behavior, mask usage, and vaccine acceptance. These association rules will be used to develop a novel modeling framework that captures spatio-temporal variations of human response to disease. The proposed modeling framework will be implemented to simulate the spread of COVID-19 using Fairfax County, VA, as a case study. This framework also will be leveraged for prescriptive analytics to find the best course of action in the event of future infectious disease outbreaks. The researchers will simulate and optimize policy measures aimed at mitigating disease spread and minimizing socio-economic impact. This optimization will combine automatic optimization tools with the expertise of researchers in policy, epidemiology, health geography, and psychology.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

疾病动态模型是重要的工具，用于预测一段时间内的病例和死亡人数，并支持决策者为传染病暴发做好准备和作出反应。然而，尽管取得了重大进展，但许多模型仍然缺乏对人类行为和流动性的现实表现，而人类行为和流动性是疾病传播的关键驱动因素。如果不考虑人类行为的复杂性，模型做出准确预测的能力就会受到限制，尤其是在较长的时间范围内。该项目研究了在疾病传播模型中包含现实的人类行为和流动性，以便1)更好地解释人类应对疾病爆发的不同方式，2)改进传染病传播的预测，以及3)帮助制定最有效的缓解政策。研究人员使用公开可用的数据，以便可以在美国任何县或州快速部署模型，以预测和减轻未来传染性呼吸道疾病的爆发（例如，COVID-19、季节性流感、麻疹和天花）。这些模型可以提供更及时和准确的预测，以帮助公众、关键机构和决策者预测即将发生的情况，并为基于证据的政策制定提供支持。该项目将为早期职业研究人员提供专业发展机会，并为有抱负的科学家暑期实习项目的博士后研究员、研究生、本科生和高中生提供培训机会。研究人员将使用数据驱动的方法来解释对疾病爆发的行为反应的时空变化。他们假设，平均收入、年龄、政治倾向等区域变量与行为反应的空间模式有关，并将利用非常大的数据集来挖掘这些变量与观察到的行为反应（包括社交距离、居家行为、口罩使用和疫苗接受）之间的关联规则。这些关联规则将用于开发一种新的建模框架，以捕获人类对疾病反应的时空变化。拟议的建模框架将以弗吉尼亚州费尔法克斯县为案例研究，用于模拟COVID-19的传播。这一框架还将用于说明性分析，以便在未来发生传染病暴发时找到最佳行动方案。这组科学家将模拟和优化旨在减轻疾病传播和尽量减少社会经济影响的政策措施。这种优化将自动优化工具与政策、流行病学、卫生地理和心理学研究人员的专业知识结合起来。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

项目成果

Human mobility-based synthetic social network generation

基于人类流动性的合成社交网络生成

• DOI: 10.1145/3557921.3565540
• 发表时间: 2022
• 期刊: HANIMOB '22: Proceedings of the 2nd ACM SIGSPATIAL International Workshop on Animal Movement Ecology and Human Mobility
• 作者: Gallagher, Ketevan;Kotnana, Srihan;Satishkumar, Sachin;Siripurapu, Kheya;Elarde, Justin;Anderson, Taylor;Züfle, Andreas;Kavak, Hamdi
• 通讯作者: Kavak, Hamdi

PhyloView: A System to Visualize the Ecology of Infectious Diseases Using Phylogenetic Data

• DOI: 10.1109/mdm55031.2022.00051
• 发表时间: 2022-06
• 期刊: 2022 23rd IEEE International Conference on Mobile Data Management (MDM)
• 作者: M. T. Le;D. Attaway;T. Anderson;H. Kavak;A. Roess;Andreas Züfle

Factorized deep generative models for end-to-end trajectory generation with spatiotemporal validity constraints

• DOI: 10.1145/3557915.3560994
• 发表时间: 2022-11
• 期刊: Proceedings of the 30th International Conference on Advances in Geographic Information Systems
• 作者: Liming Zhang;Liang Zhao;D. Pfoser

Spatiotemporal prediction of foot traffic

• DOI: 10.1145/3486183.3490997
• 发表时间: 2021-11
• 期刊: Proceedings of the 5th ACM SIGSPATIAL International Workshop on Location-based Recommendations, Geosocial Networks and Geoadvertising
• 作者: Samiul Islam;Dhruv Gandhi;Justin Elarde;T. Anderson;A. Roess;Timothy F. Leslie;H. Kavak;Andreas Z

Using Generative Adversarial Networks to Assist Synthetic Population Creation for Simulations

• DOI: 10.23919/annsim55834.2022.9859422
• 发表时间: 2022-07
• 期刊: 2022 Annual Modeling and Simulation Conference (ANNSIM)
• 作者: Srihan Kotnana;Westfield;T. Anderson;Andreas Züfle