Infectious disease doctor and BlueDot founder's early-warning system accurately predicted where COVID-19 would spread
On December 31, 2019, the Toronto-based tech company BlueDot’s software picked up an unusual news item from Wuhan, China.
A Chinese article reported an outbreak involving 27 people with pneumonia of an unknown cause, with cases appearing to be centered around a seafood market. BlueDot founder and CEO Dr. Kamran Khan immediately noticed parallels to the SARS outbreak in 2002, which started in a similar type of market in the Guangdong province.
“We didn’t know at that moment that it was a coronavirus of the next pandemic, but it was sufficiently concerning to us,” says Khan, who is also a professor at the University of Toronto and an infectious disease physician at St. Michael’s Hospital.
Before 10 am, the BlueDot team had written up a synopsis and sent it to their various clients around the world, including the Public Health Agency of Canada. On January 8, BlueDot submitted its findings in the peer-reviewed Journal of Travel Medicine in it, using global airline ticketing data, the company accurately predicted the next 11 cities where the virus would spread. The following day, the World Health Organization released its first statement regarding a flu-like outbreak in China – more than a week after BlueDot’s initial response.
Launched in 2014, BlueDot is a holistic early-warning system that uses machine-learning and AI to scrape the internet for news related to potential infectious disease outbreaks.
“We monitor open source online data, listening for early murmurs of outbreaks. In many cases, we see them well before they’re reported officially by health agencies,” says Khan.
Every 15 minutes, 24 hours a day, the platform scours hundreds of thousands of sources including official reports from global health agencies, news articles, health forums and blogs. After eliminating duplicates and filtering out irrelevant information, a team of experts that includes physicians, epidemiologists, veterinarians, geographers, data scientists and ecologists reviews the findings.
Khan’s experience working as an infectious disease specialist during the SARS crisis inspired him to build BlueDot.
“If you’re going to change the course of an outbreak, you’ve got to intervene as quickly as possible,” says Khan. “What we learned during the SARS outbreak is that you may not always get information in as timely manner as you would like it. This is not to suggest anyone is withholding information, but the process can be slow.”
In addition to its early detection platform, BlueDot also has surveillance software that can track the risk of diseases spreading. It’s currently analyzing anonymous data from millions of mobile devices around the world to assess the impacts of physical distancing measures. In March, the Public Health Agency of Canada announced it was working with BlueDot to monitor the spread of COVID-19 across the country.
“There’s only so many human resources in the public health community, so what we’ve been doing is analyzing population movements across the country to understand how our populations are physically distancing, where it’s working and where perhaps messages need to be reinforced.”
Khan believes AI has enormous potential in the health-care industry, but acknowledges that it has to be developed with scientific rigour and work in tandem with robust human teams.
“AI is a tool and piece of technology. It’s not a quasi-sentient being,” says Khan. “We rely on a machine to do what it does best, which is operate around the clock without fatigue, and we allow humans to do what we do best, which is to bring our understanding of context.”
BlueDot’s first clients were public health agencies, but they’ve begun working with the private sector too, including hospitals and airlines. Khan says that frontline health-care workers need to know of potential threats while they’re treating patients, and that airlines need to prepare for disruptions so they can better protect their passengers and crew.
“We’re trying to take more of a holistic view that everybody needs to be empowered,” says Khan. “This is something that’s been on my mind for the past 17 years, since SARS. We’ve had so many events: the Ebola outbreak, Zika, MERS, the flu pandemic in 2009. COVID-19 just happened to the big one that really stopped us all. But I think deep down, we always knew something like this was going to happen.”
Nathalie Younglai: Writer/director is making make the Canadian screenscape more equitable