Researchers develop model to more accurately track infectious disease
Computer scientists find existing models used to predict epidemics could ignore significant transmission routes.
A team of researchers from Macquarie University, QUT, and CSIRO are helping to improve infectious disease modelling by including people who can be infected by germs found in the environment, long after the person carrying them has moved on.
Current models work on the assumption that disease transmission only takes place in circumstances in which an infected and a non-infected person are present in the same space at the same time.
The work is part of the Disease Networks and Mobility (DiNeMo) project led by CSIRO, Australia’s national science agency.
“For example, someone infected with an airborne disease, such as flu, will spread it through coughing or sneezing,” says co-author, Macquarie’s Professor Bernard Mans.
“The result will be tiny particles suspended in the air which can infect the next person to come into contact with them, even though the person who sneezed has left that spot.”
Professor Mans and his colleagues – Mohammad Shahzamal and Dr Frank de Hoog from CSIRO’s Data61 and Professor Raja Jurdak from QUT – found that incorporating indirect disease transmissions into models fundamentally changed, and enlarged, the patterns of diffusion.
The new networks were further moderated by estimates of decay rates – the time it took for infective particles to lose potency.
Drawing on anonymised location data, the researchers were able to simulate disease spread via indirect transmission using 56 million data points. A new and more accurate theoretical model was developed by assuming indirect infections occurred at given distances over particular periods, tweaking the model to account for varying degrees of virulence.
“We found that including indirect transmissions changed things considerably,” explains Professor Jurdak, scientific lead of the DiNeMo project.
“It strengthened existing transmission routes but also drew in individuals who were not in the existing models because they were not directly in contact with an infected person. Overall, the spread of the disease became larger and denser.
“As population numbers continue to grow and public places become more crowded it is hoped the research will help to progress existing infectious disease mapping.”
Professor Mans – who is Interim Executive Dean of Macquarie’s Faculty of Science and Engineering – adds that the theoretical model was then verified by comparing it to the patterns in real-world disease spread.
The new model can also be used to describe non-medical spreading phenomena, such as the spread of computer viruses. The research thus has implications for areas such as cybersecurity as well as public health.
In a paper published in the journal Royal Society Open Science, the researchers add that their approach to mapping infection networks requires further research to be fully representative of real situations.
“It would be interesting to know how these higher order interactions such as people encountered or actions taken also effect transmission dynamics,” says Professor Mans.
The paper is available here: http://dx.doi.org/10.1098/rsos.190845
Diagram credit: Md Shahzamal