PIPP Phase I: An End-to-End Pandemic Early Warning System by Harnessing Open-source Intelligence

当前的全球COVID-19疫情凸显了为大流行做好准备的必要性。早期发现和尽可能的预测是关键。虽然在地球下一次大流行之前获得尽可能多的早期预警至关重要，但下一次全球疫情绝对不可能完全重演当前的COVID-19危机。开放来源情报对于提供大流行病早期预警至关重要。疾病，特别是传染病，是社会生物学现象，会留下社会和微生物的足迹。这个流行病预防预测情报（PIPP）第一阶段：发展资助项目将探索开发一个开源情报系统的可行性，该系统由来自不同来源的生物、环境、社会经济、行为和媒体数据提供信息，以监测人类社会，寻找反映具有大流行潜力的新型病原体出现的异常活动的迹象。这项多学科研究将连接生物学、社会科学、流行病学和计算机科学，以应对这一重大挑战，并开始构建一个半自主系统，对下一次大流行进行早期预警。PIPP第一阶段项目将开发一个端到端的流行病预警系统原型，该系统由人工智能（AI）、机器学习、数据科学和开源技术提供支持，可以同时寻找新发传染病或已知疾病的迹象，预测其传播，并检测和监测空间和时间上的风险因素。该系统将把人类智能作为一个综合组成部分，并可转移和扩展，以支持其他数据源和机器学习模型，使其适合于检测和预测新疾病的爆发，无论其性质和规模如何。第一阶段的活动包括四个协同试点项目，突出了应对重大挑战的创新方法，以及一项发展社区和全面中心能力的详细计划。该奖项由流行病预防的跨部门预测情报第一阶段（PIPP）计划提供支持，该计划由生物科学（BIO）、计算机信息科学与工程（CISE）、社会、行为和经济科学（SBE）和工程（ENG）委员会联合资助。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。

The ongoing global COVID-19 outbreak highlights the need to prepare for pandemics. Early detection, and, to the extent possible, prediction, are the key. While it is crucial to get as much early warning as possible in advance of the planet’s next pandemic, the odds are overwhelmingly against the next global outbreak being an exact repeat of the current COVID-19 crisis. Open source intelligence (OSINT) is vital to the provision of pandemic early warning. Diseases, especially infectious diseases, are socio-biological phenomena and leave both social and microbiological footprints. This Predictive Intelligence for Pandemic Prevention (PIPP) Phase I: Development Grants project will explore the feasibility of developing an open-source intelligence system, informed by biological, environmental, socio-economical, behavioral, and media data from diverse sources, to monitor human society for signs of unusual activities that reflect the emergence of novel pathogens with pandemic potential. This multidisciplinary research will bridge biology, social sciences, epidemiology, and computer science to address this grand challenge and start construction of a semi-autonomous system to give an early warning of the next pandemic.The PIPP Phase I project will develop a prototype of an end-to-end pandemic-early-warning system powered by artificial intelligence (AI), machine learning, data science, and open-source technologies, which can simultaneously look for signs of an emerging infectious disease or a known disease, predict its spread, and detect and monitor risk factors over space and time. The system will embrace human intelligence as an integrated component and will be transferable and extensible to support additional data sources and machine learning models, making it suitable for detecting and predicting outbreaks of a new disease regardless of its nature and scale. The Phase I activities include four synergistic pilot projects highlighting innovative approaches to addressing the grand challenges, as well as a detailed plan to develop communities and capacity of a full center.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.