Spring arrives each year in Ontario with fresh concerns over diseases such as West Nile and Lyme, which flare up during warmer weather. If we could predict where and when an outbreak might occur, it would almost certainly make it easier to manage or prevent it. This is the Holy Grail of epidemiology, a cornerstone of public health that traces its roots to the Greek physician Hippocrates and now concerns people such as Isaac Bogoch, a U of T professor in the Faculty of Medicine and a tropical disease specialist at Toronto General Hospital.
A big challenge for epidemiologists is to identify the conditions under which a disease might spread. For certain viral diseases transmitted by tropical mosquitoes, Bogoch and his research group have developed a novel approach that combines the effects of climate change with data on airline travel and the distribution of mosquitoes that carry the disease. Using this model, his team has correctly predicted the extent of two recent, high-profile outbreaks.
In 2014, they identified the likely origin and potential spread in the Caribbean of Chikungunya, a mild African viral disease carried by mosquitoes. Using arrival and departure information from Caribbean airports, the model correctly predicted how the subsequent epidemic would affect 25 Caribbean islands, northern South America and Florida – all adopted homes of the mosquitoes that transmit the virus. Within six months, more than 16,000 confirmed and 900,000 suspected cases were logged, many of which might have been prevented if regional public health officials had been provided with the resources to act on the scientists’ predictions.
As the Chikungunya epidemic waned in the spring of 2015, Bogoch and his group began modelling the potential outbreak of another African virus spread by mosquitoes: Zika. Their study, published early the following year, showed the virus’s likely transmission patterns. Again, the study predicted the epidemic accurately: Zika raged through 50 countries in the Americas, infecting millions and prompting the World Health Organization to declare a global emergency.
As the international health community responded to the outbreak, Bogoch’s group developed a model for African and Asia-Pacific countries susceptible to Zika’s spread. They identified regions and times where the health, economic and social effects from Zika would be greatest so that public health authorities in resource-limited areas could be ready to respond should the disease appear.
The utility of such modelling is clear – as is the takeaway. “The reasons for these large-scale outbreaks are increased globalization, greater individual mobility and changing weather patterns – none of which are going away,” says Bogoch. “Increasingly, it isn’t about a new virus outbreak, but trying to figure out how those we already know about will expand into new territories.”
By bringing artificial intelligence into chemistry, Prof. Aspuru-Guzik aims to vastly shrink the time it takes to develop new drugs – and almost everything else