PIPP Phase I: Heterogeneous Model Integration for Infectious Disease Intelligence

PIPP 第一阶段：传染病情报的异构模型集成

• 批准号: 2200158
• 负责人: John Drake
• 金额: $ 100万
• 依托单位: University of Georgia Research Foundation Inc
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2022
• 资助国家: 美国
• 起止时间: 2022-08-01 至 2025-01-31
• 项目状态: 未结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2200158&HistoricalAwards=false
• 关键词: PIPP; Phase; Heterogeneous; Model; Integration

During outbreaks of emerging infectious diseases, leaders in government and the private sector must make timely decisions to control spread and mitigate damages. Computer models are often used to support decision-making, for instance to provide forecasts of future transmission or to understand how public policies may be received by the population. However, infectious disease intelligence suffers from the fact that infectious disease modeling is largely disjointed because most models examine only one or two aspects of transmission, whereas epidemics and pandemics are complicated, multi-faceted events that touch on many aspects of society. For this reason, intelligence for pandemic prediction and prevention must be a multidisciplinary endeavor that integrates diverse theories, concepts, and frameworks from the natural and social sciences, although a framework for this integration is currently lacking. To fill this gap, this project will develop a Systems-of-Systems (SoS) framework that allows the knowledge gained from different disciplinary approaches to be integrated with one another. Anticipated outcomes of this framework include improved situation awareness, real-time forecasting, risk analysis, public policy interventions, and individual decision-making.The SoS modeling approach will address the grand challenge of modeling interdependence across scales by allowing for interactive feedback through the integration of information from different sources. This project will plan and execute six Demonstration Projects (DPs) that individually connect at least two different scales and/or scientific methodologies. All DPs will focus on Highly Pathogenic Avian Influenza (HPAI) as a model for the spillover and emergence of an emerging respiratory pathogen of animal origin. DP1 will develop an automated reasoning engine -- a kind of Artificial Intelligence (AI) that learns and reasons from a comprehensive, multi-disciplinary representational ontology of knowledge about HPAI across disciplines. DP2 will develop prototype explainable AI algorithms to generalize our understanding of how vaccines can be used to contain pandemics. DP3 will collect new data on the behavioral responsiveness of the population to health communications and integrate that data into disease transmission models. DP4 will develop a new dynamical model to characterize how variation in compliance with public policies creates epidemic tipping points and how these tipping points can be anticipated. DP5 will develop new methods for estimating the pandemic potential of wildlife pathogens from genetic data. DP6 will develop techniques for extrapolating the results of laboratory experiments to characterize the pandemic potential of virus lineages before they emerge in the human population.This award is supported by the cross-directorate Predictive Intelligence for Pandemic Prevention Phase I (PIPP) program, which is jointly funded by the Directorates for Biological Sciences (BIO), Computer Information Science and Engineering (CISE), Social, Behavioral and Economic Sciences (SBE) and Engineering (ENG).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.

在新出现的传染病暴发期间，政府和私营部门的领导人必须及时做出决定，以控制传播和减轻损害。计算机模型经常被用来支持决策，例如，提供对未来传播的预测，或了解公众可能如何接受公共政策。然而，传染病情报受到这样一个事实的影响，即传染病建模在很大程度上是脱节的，因为大多数模型只检查传播的一个或两个方面，而流行病和流行病是复杂的、多方面的事件，涉及社会的许多方面。出于这个原因，大流行预测和预防的情报必须是一项多学科的努力，整合了自然科学和社会科学的各种理论、概念和框架，尽管目前缺乏这种整合的框架。为了填补这一空白，该项目将开发一个系统(SOS)框架，使从不同学科方法中获得的知识相互结合。这一框架的预期结果包括改进的形势感知、实时预测、风险分析、公共政策干预和个人决策。SOS建模方法将通过整合不同来源的信息来允许交互反馈，从而解决跨规模相互依赖建模的巨大挑战。该项目将规划和执行六个示范项目(DP)，分别连接至少两个不同的规模和/或科学方法。所有DPS将重点关注高致病性禽流感(HPAI)，作为一种新出现的动物源性呼吸道病原体溢出和出现的模型。DP1将开发一个自动推理引擎--一种人工智能(AI)，它从一个全面的、多学科的、关于跨学科的HPAI知识的代表性本体中学习和推理。DP2将开发原型可解释的人工智能算法，以推广我们对疫苗如何用于遏制流行病的理解。DP3将收集关于人群对健康传播的行为反应的新数据，并将这些数据整合到疾病传播模型中。DP4将开发一个新的动态模型，以表征公共政策遵从性的变化如何产生流行病临界点，以及如何预测这些临界点。DP5将开发新的方法，根据遗传数据估计野生动物病原体大流行的可能性。DP6将开发外推实验室实验结果的技术，以在病毒谱系出现在人类种群之前表征大流行的潜在特征。该奖项由跨部门大流行预防第一阶段预测情报(PIPP)计划支持，该计划由生物科学(BIO)、计算机信息科学和工程(CEISE)、社会、行为和经济科学(SBE)和工程(ENG)理事会共同资助。该奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。