RAPID: Accelerated Testing for COVID-19 using Group Testing

新冠肺炎引发了一场史无前例的全球健康危机，在接下来的几个月里，这场危机可能会变得更加普遍。众所周知，对有症状和无症状的人进行广泛和立即的检测对于实施遏制政策和确保医疗资源可以适当分配到不同的地理区域非常重要。个人检测可以提供必要的信息；然而，这需要大量的医疗和人力资源。该项目将促进新冠肺炎的广泛测试，同时使用更少的测试。主要的方法是基于汇集多个患者的样本并对组合样本进行测试的想法。如果检测结果为阴性，则可以得出池中没有人被感染的结论，如果结果为阳性，则可以执行进一步的细粒度检测。基于汇集的检测，也称为群体检测，可以非常有效地减少识别人群中的感染者和获得关于感染率的粗粒度人口水平信息所需的检测次数。在这个项目中，将设计有效的分组测试方案，以最大限度地减少所需的测试总数和/或进行测试所需的总时间，并对其性能进行分析。在这个项目中，将设计和优化不需要准确了解感染率的群体检测计划和多阶段群体检测计划。还将描述所需测试次数和完成测试所需的总时间之间的权衡。将研究分组测试方案对被测试人群中感染状态的相关性和测试中的错误的稳健性。利用群体检验和假设检验的数学工具，还将设计和分析对感染率进行快速分类的策略。最后，将开发一个智能手机应用程序，指导实验室技术人员完成集体测试过程。重点将放在小池规模和人口规模上。成功完成该项目中建议的活动将推动组测试领域的发展，并为新冠肺炎测试提供实用和高效的解决方案。该奖项反映了美国国家科学基金会的法定使命，并已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。

COVID-19 has resulted in an unprecedented global health crisis that may become even more widespread over the upcoming months. Extensive and immediate testing of symptomatic and asymptomatic people is known to be important for implementing containment policies and to ensure that medical resources can be apportioned to different geographic regions appropriately. Individual testing can provide the necessary information; however, this requires enormous amounts of medical and human resources. This project will facilitate widespread testing for COVID-19 while using fewer tests. The main approach is based on the idea of pooling samples from multiple patients and performing tests on combined samples. If the result of a test is negative, one can conclude that no one in the pool is infected, and if the result is positive, then further fine-grained testing can be performed. Pooling-based testing, also known as group testing, can be very effective in reducing the number of tests required for both identifying infected people in a population and for obtaining coarse-grained population-level information about infection rates. In this project, effective group-testing schemes that minimize the total number of tests required and/or the total time taken to conduct tests will be designed, and their performance will be analyzed. In this project, group-testing schemes that do not require precise knowledge of the infection rates and multi-stage group-testing schemes will be designed and optimized. The trade-off between the number of tests required and the total time taken to complete testing will also be characterized. The robustness of group-testing schemes to correlation in the infection status among the tested population and to errors in the tests will be studied. Using mathematical tools from group testing and hypothesis testing, strategies for rapid classification of infection rates will also be designed and analyzed. Finally, a smart-phone application which guides laboratory technicians through the group-testing process will be developed. The focus will be on small pool sizes and population sizes. Successful completion of the proposed activities in this project will advance the state of the art in the field of group testing, and provide practical and efficient solutions for COVID-19 testing.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.