Since it was first identified in the city of Wuhan (China), the COVID-19 coronavirus has infected over 121,000 people, killing over 4,000 patients. While medical scientists work 24/7 to find a vaccine, Artificial Intelligence (AI) researchers have joined the global effort to defeat the virus.
Forecasting the outbreak
According to an article published by CNBC (1), the first signs of the outbreak were picked up by Canadian AI platform BlueDot. Shortly after midnight on December 30, 2019, the platform started registering an unusual spike in pneumonia cases around a market in Wuhan. “We didn’t know at that moment that this was going to become something of this magnitude,” says Kamran Khan, founder and CEO of BlueDot and professor of medicine and public health at the University of Toronto. The idea of building a platform capable of tracking, locating and conceptualizing infectious diseases, came to Khan’s mind following the 2003 SARS outbreak. BlueDot employs language processing to gather data from hundreds of thousands of sources worldwide, including statements from public health organizations, population demographics, livestock health reports, social media, and airline ticketing platforms. This information is then analyzed by a team of physicians and programmers, who compile reports and send them to healthcare, government, public health and private clients. The aim is to create an early warning system that will allow governments to keep one step ahead of the virus.
In the case of COVID-19, BlueDot was able to predict where the virus would have spread to by analyzing global airline ticketing data. The international destinations that the platform forecasted would receive the highest volume of travelers from Wuhan were Bangkok, Hong Kong, Tokyo, Taipei, Phuket, Seoul, and Singapore. Eventually, eleven of the cities at the top of BlueDot’s list were the first places where COVID-19 emerged.
Tracking the contagion
With the outbreak now spreading throughout the planet (with over 121,000 confirmed cases) the team at Boston Children’s Hospital is keeping constantly updated a map which allows to live track the virus (https://www.healthmap.org/covid-19/). Even in this case, the researchers are using machine learning and the analysis of big data to collect and examine information from open sources available online. The machine’s algorithm is capable of scanning social media posts mentioning symptoms resembling respiratory problems and fever. "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 Dr. John Brownstein, chief innovation officer at Boston Children's Hospital (2). Once information which could point to a possible outbreak in an area is found, the team shares them with other health organizations worldwide. It’s a race against time to defeat the virus and save lives. Technology and AI could accelerate our sprint towards victory.
Other applications of predictive analysis
The team at Grant Thornton is able to apply predictive analytics and machine learning techniques to predict behaviours and patterns. In a business environment, this gives organizations the possibility of forecasting future performance by analyzing past events. This is paired with industry knowledge and experience that complements any data-driven strategy. Our clients represent a diverse background of businesses, including retail, hospitality services, financial services, real estate, and iGaming. Get in touch with us to unleash the power of big data!
1 - https://www.cnbc.com/2020/03/03/bluedot-used-artificial-intelligence-to-predict-coronavirus-spread.html
2 - https://abcnews.go.com/Health/doctors-artificial-intelligence-track-coronavirus-outbreak/story?id=69444963