January 9 World Health Organizationnotified the public of an outbreak of coronavirus in China. As of January 27, the official number of infected people is at least 2,700 people, and the death toll is in the tens. However, the CDC, the US Centers for Disease Control and Prevention, released information on the outbreak of the dangerous 2019-nCov a few days earlier, on January 6. Even more surprising is the reported coronavirus outbreak of the Canadian Health Monitoring Platform on December 31, 2019. But how did they succeed?
How does Artificial Intelligence predict epidemics?
We often talk about what scientists aroundthe world is concerned about the growing threat of an epidemic of a particular infection. Climate change makes the planet warmer, and this allows insects such as mosquitoes - which are carriers of malaria and Zika fever - to move to new regions in which it was much colder before. However, mosquitoes are not the only threat. So, recently, scientists have discovered a whole bunch of ancient viruses that have been locked in ice for 15 thousand years. And meat consumption, in turn, increases the number of animals on farms, which allows viruses to constantly mutate.
Read even more interesting articles about what is happening in the world of popular science and high technology on our channel in Yandex.Zen
For this reason, scientists today useAI-driven BlueDot algorithm that looks at news of animal and plant diseases in foreign languages, as well as official statements to warn the public in advance of not visiting hazardous areas such as Wuhan. In the case of coronavirus, the 2003 experience is important - during the SARS outbreak, when the Chinese government for a long time hid information about the epidemic. And the speed of response during outbreaks of infectious diseases is extremely important.
According to the Wired publication, BlueDot was launched in2014 and attracted venture financing in the amount of 9.4 million dollars. Currently, the company has 40 employees - doctors and programmers, who are developing an analytical disease surveillance program that uses natural language processing and machine learning methods to view news in 65 languages, as well as airline data and outbreaks of animal diseases . The developers clarify that the algorithm does not take into account publications on social networks, since this data is too messy.
Do you think this algorithm will be successful in predicting the spread of infections in the future? Let's discuss AI capabilities in our Telegram chat and in the comments on this article.
After automated data samplingcompleted, the findings of the algorithm are checked by epidemiologists. The report is then sent to public health officials in dozens of countries (including the US and Canada), airlines, and hospitals where infected patients may be. So, in the case of distribution, the BlueDot algorithm was able to correctly predict that the coronavirus from Wuhan will appear first in Bangkok, then in Seoul, Taipei and Tokyo. Moreover, in the past, the algorithm successfully predicted an outbreak of Zika virus in South Florida.