covid model
Original Published Date: 
Tuesday, March 30, 2021

Full article republished by The University of Melbourne from an original in The Conversation.

Researchers at the ARC Centre of Excellence for Mathematical and Statistical Frontiers of Big Data, Big Models, New Insights (ACEMS), with support from a Discovery Projects grant, have designed a safe ‘virtual’ epidemic, to help us learn about the spread of COVID.

A joint project between researchers at The University of Queensland, The University of Melbourne, Macquarie University, and several international institutions, the epidemic model uses virus-like tokens, that are spread between mobile devices via Bluetooth, similarly to how a biological virus spreads between people. Called 'Safe Blues', the tool could help public health authorities better control outbreaks by providing data on the general level of contact between people.

Safe Blues’ machine learning methods have been developed and evaluated using mathematical simulation models, and researchers say that initial results show an ensemble of Safe Blues token strands can yield powerful estimates of actual epidemic behaviour.

The research team is now working towards a system pilot at The University of Auckland. Using an experimental Safe Blues Android app, the aim will be to generate and study how the virtual virus spreads in a campus setting.

For more prediction power, the researchers propose that insights provided by the app could also be coupled with data from waste water measurements and from existing social networks such as Google, Apple or Facebook.

Photo credit: 

Image: Pixabay (Public Domain).