A global platform with medical data helps in the diagnosis of unknown viruses. However, it cannot prevent the spread of the coronavirus. To this end, the appropriate actors must exchange information in good time.

It is a catastrophe that is of concern to the whole world: the coronavirus has already caused more than 1100 deaths and more than 44,000 cases are known. It spread rapidly across national borders. And in Germany, too, as of March 2020, according to the US Johns Hopkins University, there are now more than 2000 proven infections. This is a manageable number, but it is still a cause for concern in many places. At the beginning of the pandemic, the diagnosis of the novel pathogen discovered in Wuhan was about as difficult as that of swine flu some time ago. Many things would be easier and also unnecessary quarantine periods avoided or at least shorter if the lung specialist had the possibility to compare the virus sample of a cough with millions of other samples in a global medical database in order to determine whether it is an unknown virus. The US company SAS Institute is actually providing such a database, which is currently being used by pharmaceutical companies and clinics for in-house analysis. But it would also have what it takes to read out health data worldwide. However, researchers have now discovered that, depending on the region, data from social networks is much more meaningful.

Exchange information faster and react accordingly

This is exactly what the Canadian company BlueDot did: blog entries about the coronavirus were read out in 65 languages by means of AI. Access to global airline ticket data made it possible to predict how the virus would spread. The automatic results were then scientifically tested by epidemiologists. With this human-monitored AI system, BlueDot provides the current gold standard for early warning systems of pandemics. Thus it could be correctly predicted that the corona virus will first spread towards Bangkok, Seoul, Taipei and Tokyo. Of course, this has not been able to prevent the spread. To this end, it is important that governments exchange information more quickly and react accordingly.

Which airports should currently be avoided is again shown by the risk monitor, which was compiled by scientists from the Humboldt University of Berlin and the Robert Koch Institute. Based on the data of thousands of flight connections, it shows the probability that a guest from Wuhan has arrived at one of 1900 airports worldwide. The front-runner on our continent is Charles de Gaulle Airport in Paris with two out of every 1000 passengers taking off from faraway China. German airports are in midfield with an average of 1.4 persons.

* This text may contain translation errors as the translation was performed by an online translation tool.