RAPID: Collaborative: A privacy-preserving contact tracing system for COVID-19 containment and mitigation

与新冠肺炎抗争的一个关键工具是一个接触者追踪系统，该系统可以识别过去与确诊病例有过密切接触的个人。这样的机制可以向这些人发出警报，使他们能够自愿自我隔离。此外，当这些人开始出现症状时，接触者追踪系统可以优先对他们进行检测，这将更有效地利用有限的检测能力，并为受感染的个人提供早期治疗。然而，由于美国和许多西方国家严格的隐私保护法律，部署这样的系统是一个挑战。为了解决这一亟待解决的问题，本项目构建了一个隐私保护的接触者追踪系统，名为COVID检测器。COVID检测器依靠智能手机来跟踪患者和健康人的过去位置，并利用加密计算来确保在接触者跟踪计算期间不会暴露私人数据。该项目满足了在一个现代、复杂和高度联系的世界中以隐私友好的方式支持用户联系人追踪的更广泛的需要。目前，在平衡公共健康和个人用户隐私的背景下，这一讨论具有很高的相关性。虽然现有的新冠肺炎接触者追踪应用程序很少，但用户/患者身份的匿名性是它们能够提供的最好的保护用户隐私的方法。COVID探测器是第一个同时为用户/患者身份和用户/患者轨迹数据提供强大保护的探测器。冠状病毒检测器使用同态加密技术，在密文域将确诊的新冠肺炎患者的轨迹与健康用户的轨迹进行匹配，以便健康用户确定他们的感染风险级别。匹配过程既不会向任何一方透露健康用户的位置数据，也不会向任何一方透露患者的位置数据。在轨迹跟踪中使用同态加密技术是新颖的，以前从未应用过。这一奖项反映了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。

A crucial tool in the fight with COVID-19 is a contact tracing system that can identify individuals who had close contacts with confirmed cases in the past. Such a mechanism can alarm these individuals so that they can voluntarily self-quarantine. In addition, when these individuals start to have symptoms, a contact tracing system can prioritize them for testing, which will make more efficient use of the limited test capacity and provide early treatment for infected individuals. However, due to strict privacy-protection laws in US and many western countries, it is a challenge to deploy such a system. To solve this urgent problem, this project builds a privacy-preserving contact tracing system, named COVID Detector. COVID Detector relies on smartphones to track both patient and healthy person's past locations and leverages cryptographic computations to ensure that no private data is exposed during the contact tracing computation. This project addresses the broader need to support, in a privacy-friendly manner, user contact tracing in a modern, complex and highly-connected world. This discussion is presently highly relevant in the context of balancing public health and individual user privacy.While there are a few existing contact tracing apps for COVID-19, anonymity of user/patient identity is the best that they can provide to protect user privacy. COVID Detector is the first that provides strong protection of both user/patient identity and user/patient trajectory data. COVID-Detector uses homomorphic encryption techniques to match the trajectory of confirmed COVID-19 patients with that of healthy users in the ciphertext domain, so that healthy users can determine their infection risk level. The matching process reveals neither healthy users' nor patients' location data to any party. The use of homomorphic encryption technologies in trajectory tracking is novel and has not been applied before.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.