Zombies have bitten into popular culture over the last decade.
Brian Blais, professor of science and technology at Bryant University, examines where the fiction ends and the reality of disease epidemics begins.
Dr. Brian Blais received his Bachelor’s in Physics at Wesleyan University and his PhD in Physics at Brown University studying mathematical models of the brain under his long-time mentor and collaborator Nobel Laureate Dr. Leon Cooper. He has worked at Bryant University, in Rhode Island, for the past 17 years studying topics in computational neuroscience, statistical inference, and dynamical systems. In addition he has worked to use popular culture to help motivate and inspire learning, demonstrating his passion for bringing technical scientific topics to everyone.
Zombies, Influenza and other diseases
An epidemic of zombies has sparked public interest for many years, from the dark-comedy Shaun of the Dead, blockbuster movies like World War Z, up to the recent TV series The Walking Dead. While entertainment, these fictional depictions of epidemics have direct parallels with real-world diseases like ebola and influenza. To see this parallel, we can apply the same mathematical models used to predict the dynamics of influenza epidemics. This is also a useful exercise in model building, or in other words, how do you map a mathematical model to the real-world? A common model of diseases divides the total population into three sub-populations referred to as Susceptible, Infected, and Recovered. The Susceptibles don’t have the disease yet, but could get infected by interacting with the Infected. In the case of influenza, this interaction occurs through the exchange of water droplets carrying the virus, but in the case of zombies it happens with a zombie bite. The entire process can be written mathematically, and simulated with various infection rates, incubation periods, and recovery rates using known processes of the virus. But, with zombie epidemics – and many real-world epidemics – the only data we have is the total number of infected individuals over time. Thus, we need a process to infer the infection rates from this kind of data alone. The process we can employ is referred to as Bayesian parameter estimation. It provides a framework for comparing different mathematical models of the world and supplies estimates of the unknown parts of the models, like infection and recovery rates. Applying this to many different zombie stories, we find that, surprisingly, the zombie virus is actually less infectious than the measles. Thus, as a society, we should probably be concerned more about a possible measles epidemic than the next zombie apocalypse.
Read More:
Professor’s Webpage
Cornell University Library – Bayesian Analysis of Epidemics – Zombies, Influenza, and other Diseases
