Facing this tremendous global pandemic, it is clear that governmental institutions have fallen very short. On behalf of all concerned professionals and academics globally, our organization is working to fill the gaping technological void in this response. To meet this challenge, we are developing a scaleable API framework, SafeTrace. This decentralized database enables the widespread adoption of contact tracing via mobile devices. With this technology, the world can trace the coronavirus outbreak at the level of individuals. Yet, with this compelling technology comes a serious concern of user privacy. The key distinction of our endeavor is the level of privacy SafeTrace can provide. Complete user privacy is achieved using state of the art Multi-Party Computation protocols. This whitepaper will proceed by first motivating the necessity of this technology (Section 2), followed by technical background (Section 3), and an overview of our solution (Section 4 and beyond).
Proposed Epidemiological Models and Methods toImprove COVID-19 Contact Tracing and Outbreak Prediction
In the case of pandemics—such as the current COVID-19 pandemic—disease spreads rapidly and testing resources are limited. Contact tracing provides an opportunity to fill some of the gaps left by testing limitations. Individuals with confirmed and presumed or suspected positive cases can be identified via user input of testing status or symptoms, as well as by other data sources. The location and movement of these individuals can then be traced. This location information can be used to create a 4D model of locations over time, and the data can be input into a network model to identify other individuals who were likely exposed as well as identify where or among which groups the next outbreaks are likely to occur. This paper describes a few central models based in epidemiology, graph theory, and population biology that can be used, along with location, case, and symptomatology data to create an effective contact tracing algorithm.It then proposes a SEIR model for tracing infections through a network and discusses limitations and options for further fine-tuning. Such an algorithm has immense potential to drastically improve epidemiological studies, reduce cases, and prioritize interventions during COVID-19 and future epidemics. The algorithm be implemented via the Safetrace API, part of the MutualAid.world project.
Keywords: contact tracing, COVID-19, SEIR model, network dynamics, SafeTrace API
Faction Co-Operative Report: Sensitive Data Management
Secure data storage with strong privacy guarantees is paramount. We aim to give data owners safety and full control over their Faction data, with minimal technical requirements or potential for error on the side of co-op members. We further aim to give Faction members the ability to authorize data for use in Secure Multi-Party Computation (MPC). Using the cryptographic techniques of MPC, a given program can be securely evaluated, whereby the inputs remain strictly private and only the final output of the desired computation is public (to au- thorized parties). In this way Faction data can be computed on without ever being revealed even to the machines jointly processing this data.
Secure Location Intersection
Tell if two people were near the same places and times… without knowing where else they were.