Doctors using artificial intelligence to track coronavirus outbreak
A team in Boston is using machine learning to track the virus.
With novel coronavirus spreading throughout the United States, researchers are turning to social media and artificial intelligence to track the virus as it spreads.
A team headquartered at Boston Children's Hospital is implementing machine learning to scour through social posts, news reports, data from official public health channels and information supplied by doctors for warning signs that the virus is taking hold in locations outside of China.
"There's incredible data that's locked away in various tools like online news sites, social media, crowdsourcing, data sources, that you wouldn't think of that would be used for public health," said Dr. John Brownstein, chief innovation officer at Boston Children's Hospital. "But actually they have incredible amounts of information that you wouldn't find in any sort of traditional government system."
More than 95,000 people around the world have been infected by the outbreak of novel coronavirus, and more than 3,200 have died -- most in China.
Tune into ABC News Live at noon ET every weekday for the latest news, context and analysis on the novel coronavirus, with the full ABC News team where we will try to answer your questions about the virus.
Brownstein leads the HealthMap team at Boston Children's Hospital. The team, using artificial intelligence, is looking for social posts that mention specific symptoms resembling respiratory problems and fever from areas where doctors have reported potential cases. The end result is a publicly accessible map that attempts to live-track the virus.
"Whether it's social media, online news reports, blogs, chat rooms -- we're looking for clues about symptoms, reports of disease, that tell us something unique is happening," said Brownstein.
Brownstein referred to this sort of online data as "digital bread crumbs."
"We organize that content, filter it, tag it and build what's essentially a global map of emerging infectious diseases that ultimately tells us early on when something is bubbling up in a particular country and a particular location," Brownstein told ABC News.
Brownstein's program is part of Boston Children Hospital's Innovation & Digital Health Accelerator (IDHA) division. The group actively shares imperative information about COVID-19 with health organizations around the world as they find it. The challenge is verifying the potential COVID-19 cases the team finds.
"The reality is, it's very challenging to know for sure when someone would have potentially have been exposed," said Brownstein. "And that's why we're seeing these chains of transmission that have occurred likely in Washington, because not all cases were identified."
There have been 15 total deaths from the novel coronavirus in the United States, with 14 in Washington and one in California. In the United States, at least 188 confirmed cases have been detected through the local public health system. There are an additional 49 Americans diagnosed with the virus who were either repatriated from the Diamond Princess cruise ship in Japan or were evacuated from Wuhan on a U.S. government-chartered plane.
Although the team has been tracking the virus since its inception, Brownstein says it is still too soon to tell how quickly it will spread in the United States.
"We don't really know the extent of the impact it'll have in populations in different communities around the US, and this is something that we have heightened surveillance around," said Brownstein. "It's very likely that we don't actually know all the cases that have emerged in the U.S. at this point."
Related Topics
Popular Reads
ABC News Live