How a Canadian company spotted the coronavirus outbreak before health officials
TORONTO -- A Canadian company’s advanced artificial intelligence system was among the first in the world to notice the new coronavirus emerging from China, beating international health authorities.
Toronto-based BlueDot uses AI to scour more than 100,000 articles every day in 65 languages looking for news about more than 150 different diseases. Around 10 a.m. EST on Dec. 31, their system spotted an article in Chinese about a “pneumonia of unknown cause” with 27 cases.
“We didn’t know at that moment that this would turn into what it has, but we knew it had the ingredients to become something,” Dr. Khan explained to CTVNews.ca in a phone interview.
The very same day, BlueDot issued an alert to its clients. That warning came nearly a week before the U.S. Centers for Disease Control and Prevention and the World Health Organization issued their public alerts about the novel coronavirus.
The company’s analysis of the likely destinations for the spread of the disease was published in the Journal of Travel Medicine on January 14.
“The cities up at the top of the list at highest risk were the first cities that received the coronavirus after it started to spread out of China, places like Tokyo, Hong Kong, Seoul, Taipei and Macau, et cetera,” Dr.Khan said.
HOW THEY DID IT
BlueDot conducts surveillance of infectious diseases from official sources reported by government agencies, through the public health and healthcare community, online forums and world media, Dr. Khan explained.
“The internet is an important medium for gathering unofficial information, but it’s so vast,” he said.
Once the data is collected, a team of experts at BlueDot then sift through the information and review a handful of what Dr. Khan called “needles in the haystack.”
“Once we think that the potential threat is legitimate we enter that into our platform and it automatically connects to the whole world’s air travel data,” said Dr. Khan, who also works as a professor of medicine and public health at the University of Toronto.
“Now we can analyze hundreds of millions or even billions of pieces of data and then instantly map out the places where this particular outbreak might go to next.”
Dr. Khan was starting his career as a doctor and scientist at Toronto’s St. Michael’s Hospital during the SARS outbreak in Canada, which killed 44 people.
“It was a profound event where I had a chance first hand to see a colleague of mine get SARS. Thankfully she survived,” Dr. Khan said.
Khan spent the next decade studying outbreaks with an emphasis on how they might spread through commercial air travel.
With the creation of BlueDot, the company aims to disseminate information faster than the diseases can spread. The company’s clients include Canadian government agencies, health and security departments, and businesses.
The coronavirus has now infected more than 34,576 people globally and killed more than 722, almost all of them in China.
Dr. Khan said multiple converging global forces are responsible for the spread of the virus, including population growth, urbanization, climate change and air travel.
“One of the biggest drivers is the mass consumption of animals, in particular wildlife,” he said. “These are the spill-over events where viruses in wild animals get into humans. This was the same story 17 years ago that it is now, so Mother Nature is trying to tell us something.
“The flip side of that is that we’re also in an era of increasing access to data, things like advanced analytics (and) AI to makes sense of data and digital technologies to communicate information.
“There’s never been a period in human history with the magnitude and scale of outbreaks that we have seen in the past 20 years.”
The company’s name, BlueDot, is a metaphor for our small world, according to its website.
Established in 2014, BlueDot now employs 40 people -- a mix of science, technology and healthcare experts.