A Wuhan virus victim lies on the floor of a hospital, awaiting attention. (Source: nyt.com video still)

WIRED: BlueDot Algorithm Spots Virus Threat a Week before Officials

As early as December 31, a Canadian algorithm called BlueDot spotted the coronavirus outbreak in China before any major health organization, according to a story on wired.com. The AI scans foreign-language news reports, animal and plant disease networks and official proclamations to find such trends for its clients.

The U.S. Centers for Disease Control and Prevention announced the outbreak on January 6; the World Health Organization issued a warning three days later. China has a poor track record of alerting people to outbreaks, which was what inspired the founder of the company to start it after the SARS outbreak in 2003.

“We know that governments may not be relied upon to provide information in a timely fashion,” Kamran Khan, BlueDot’s founder and CEO told WIRED. “We can pick up news of possible outbreaks, little murmurs or forums or blogs of indications of some kind of unusual events going on.”

A hospital infectious disease specialist in Toronto, Khan saw SARS kill 44 people there. In 2014, he started BlueDot with $9.4 million in venture capital. The 40 employees include physicians and programmers who developed a “disease surveillance analytic program, which uses natural-language processing and machine learning techniques to sift through news reports in 65 languages, along with airline data and reports of animal disease outbreaks.” Epidemiologists review the information, analyze it and form projections. BlueDot predicted the Zika outbreak in Florida.

The AI service delivers its reports to public health officials in a dozen countries (including the U.S. and Canada), airlines and some major hospitals.

The death toll from the Wuhan virus has risen to 106 in China, has infected more than 2,800 people and has spread to at least 14 other countries, according to a report on businessinsider.com on January 27. Health experts who conducted a simulation three months ago say that without early detection, their computer model had predicted that an unchecked spread of such a virus could kill 65 million people in a pandemic. Already, scientists using AI-generated models predict the outbreak will become much worse than it is now.

read more at wired.com