RAPID: Information Retrieval and Graph Mining Techniques to Enable Self-Assessment and Clinical Monitoring of Emergent Infectious Diseases

该项目将开发一个系统，使罗得岛的患者能够输入并监测与COVID-19相关的可能症状。该系统还将使患者能够将他们的症状与该州的其他人进行比较。该系统最终将提供自动化指导，指导个人是否应该（1）寻求立即的临床评估;（2）在家管理症状;或（3）继续监测症状。该系统将成为一个门户网站的一部分，该门户网站使患者可以从整个罗得岛医疗保健系统的医疗提供者获得临床数据。来自该系统的数据将用于开发计算机模型，可用于更好地预测潜在的后续COVID-19疫情。由此产生的系统和计算机模型将被设计用于更好地预测未来其他潜在的传染病爆发。该RAPID项目将专注于数据科学和信息学方法的发展和进步，以建立一个急需的基础设施，支持人群水平的综合征监测，最初的重点是支持患者查询和管理医疗系统上的患者负荷。具体而言，该项目的目标是开发一个基于患者报告和电子健康记录（EHR）数据的实时监测系统，以促进公共卫生应对COVID-19等新出现的流行病。具体而言，该小组将：（1）创建一个以患者为中心的系统，更好地通知适当的医疗服务寻求;（2）开发一个人口级临床决策支持和监测平台，使用电子病历系统，使用来自罗得岛各地医疗服务提供者的数据。该系统将提供与健康状况的一致性，可将其与资源联系起来，以指导后续建议。焦点小组将被用来探索面向患者的系统的可接受性和可行性。然后，将根据入院、出院和转院的数据以及临床和实验室数据生成一个概率性疾病状况模型，用于指导向感染风险最高的个人发出有针对性的警报。该模型还将包括时间信息，以便能够对进展速率和入院诊断时间进行建模。该模型的有效性将由一组临床医生进行评估。该模型将用于确定疾病进展的模式，以及确定公共卫生干预的目标，如地理热点或其他风险社区。该项目的成功有可能对罗得岛应对COVID-19产生直接影响，并有可能通过在全国范围内使用EHR为疾病监测流程提供信息。该奖项反映了NSF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。

This project will develop a system for enabling patients in Rhode Island to enter and monitor possible symptoms associated with COVID-19. The system will also enable patients to compare their symptoms to others in the state. The system will eventually provide automated guidance on whether and individual should (1) seek immediate clinical assessment; (2) manage symptoms at home; or (3) continue monitoring symptoms. The system will be part of a web portal that makes clinical data available to patients from medical providers across the Rhode Island healthcare system. Data from the system will be used to develop computer models that can be used to better predict potential subsequent COVID-19 outbreaks. The resulting system and computer models will be designed to better predict other potential infectious disease outbreaks in the future. This RAPID project will focus on the development and advancement of data science and informatics approaches to establish a much-needed infrastructure for supporting population-level syndromic monitoring, with an initial emphasis on supporting patient inquiry and managing patient loads on the healthcare system. Specifically, the goal of this project is to develop a real-time monitoring system based on patient-reported and electronic health record (EHR) data to facilitate public health responses to emerging epidemics like COVID-19. Specifically, the team will: (1) Create a patient-centered system that better informs appropriate seeking of healthcare services; and (2) Develop a platform for population-level clinical decision support and monitoring that uses data from healthcare providers across Rhode Island using an EHR system. The system will provide an alignment to health conditions that can be linked to resources to guide follow up recommendations. Focus groups will be used to explore acceptability and feasibility of the patient-facing system. A probabilistic symptom-condition model will then be generated from a combination of data from admissions, discharges and transfers, as well as clinical and laboratory data, which will be used to guide targeted alerts to individuals at highest risk of infection. This model will also include temporal information to enable modeling of the rate of progression and time to admitting diagnosis. The validity of the model will be assessed by a team of clinicians. The model will be used to identify patterns of disease progression, as well as identify targets for public health intervention such as geographical hotspots or other at-risk communities. The success of this project has the potential to have immediate impact in Rhode Island in its response to COVID-19, and potentially inform disease monitoring processes through the use of EHRs nationally.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.