Developing algorithms that use individual mobility data to support tracking and contact tracing of SARS-CoV-2 causing COVID-19 and future pandemics

开发使用个人移动数据的算法来支持对导致 COVID-19 和未来大流行的 SARS-CoV-2 进行追踪和接触者追踪

• 批准号: 552682-2020
• 负责人: Friesen, Marcia
• 金额: $ 3.64万
• 依托单位: University of Manitoba
• 依托单位国家: 加拿大
• 项目类别: Alliance Grants
• 财政年份: 2020
• 资助国家: 加拿大
• 起止时间: 2020-01-01 至 2021-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=703925
• 关键词: Developing; algorithms; use; individual; mobility

This proposed work develops technology that can ultimately support public health by drawing on the opportunities inherent in machine learning and modeling. Specifically, this work proposes to develop algorithms that can support the ability to track individuals during COVID-19, future pandemics, and other states of emergency. These algorithms will not function independently like a smartphone app. Instead, it will be designed to work within a larger software platform that can leverage data from multiple sources (telecommunication service providers, social media apps, financial data) and demonstrate the trajectories of both individuals and groups on fine (neighbourhood) and coarse (regional) scales in disease spread and mitigation models. When public health tools such as quarantine, isolation, and physical distancing measures are applied, the algorithms could support efforts to identify and isolate infected individuals as early as possible, carry out comprehensive contact tracing of known infected individuals, and visualize mobility patterns in communities to assess the most effective community-based strategies to minimize infection transmission (e.g. various levels of 'lockdown' and easing of restrictions/return to 'normal'). Our partner, Sightline Innovation (SI), is a Canadian AI company. SI has developed a data trust platform that provides a governance structure around enterprise use of applied machine learning. The proposed work will provide SI with an opportunity to evaluate the capability of their data trust platform, specifically evaluating additional capacities that the project offers to their existing technology-assisted disease modeling and contact tracing applications toward the COVID-19 and future pandemics. The algorithms that result from this project will ultimately be made available to any entity with interests in incorporating mobility data for disease spread and mitigation modeling.

这项拟议的工作开发的技术，可以通过利用机器学习和建模中固有的机会，最终支持公共卫生。具体而言，这项工作提出开发算法，以支持在COVID-19、未来大流行和其他紧急状态期间跟踪个人的能力。这些算法不会像智能手机应用程序那样独立运行，而是被设计成在一个更大的软件平台上工作，该平台可以利用来自多个来源（电信服务提供商、社交媒体应用程序、金融数据）的数据，并在疾病传播和缓解模型中展示个人和群体在精细（社区）和粗略（区域）尺度上的轨迹。当应用检疫、隔离和物理距离措施等公共卫生工具时，这些算法可以支持尽早识别和隔离感染者，对已知感染者进行全面的接触者追踪，并可视化社区的流动模式，以评估最有效的基于社区的战略，以尽量减少感染传播（例如，不同程度的“封锁”和放宽限制/恢复“正常”）。我们的合作伙伴Sightline Innovation（SI）是一家加拿大人工智能公司。 SI开发了一个数据信任平台，提供了一个围绕企业使用应用机器学习的治理结构。 拟议的工作将为SI提供一个评估其数据信任平台能力的机会，特别是评估该项目为现有技术辅助疾病建模和接触者追踪应用提供的额外能力，以应对COVID-19和未来的大流行病。 该项目产生的算法最终将提供给任何有兴趣将移动数据用于疾病传播和缓解建模的实体。