Proximity-based contact tracing relies on user device interaction to estimate the spread of disease. ShareTrace is one such approach that has been shown to provide improved efficacy in tracking the spread of disease by also considering (in)direct forms of contact. In this work, we aim to provide an efficient and scalable formulation of ShareTrace by utilizing asynchronous, parallel, non-iterative message passing on a temporal graph. We also introduce a unique form of reachability, message reachability, that accounts for the dynamic nature of message passing and the temporal graph. Our evaluation on both synthetic and real-world temporal graphs indicates that correct parameter values optimize for accuracy and efficiency. In addition, we demonstrate that message reachability can accurately estimate the impact of the risk of a user on their contacts.
View on arXiv