How Can AI Help Preventing A Pandemic Such As Coronavirus?


Wuhan’s coronavirus numbers continue to grow. The 2019-nCov continues to be confirmed in more individuals as the deceased keep increasing. However, the progression of those affected follows the calculated expectations.

Moreover, it does not seem to have gone out of control. The number of majorly infected regions continues to be limited to China and the nearest areas. Could the virus have been prevented from spreading across the region? Is there any tool to stop its progression further ahead? Yes, AI.

These are the stages to be followed by AI:

First Phase: Predict

This phase is prior to everything we are living in. It basically consists of generating data, investigating and applying models to predict potential pathogens and their behavior. We know what our main enemies are and we use that knowledge to be prepared, have a battery of data and prevent how a potential situation would develop.

An algorithm predicted how the virus would spread by jumping from the Chinese region to Bangkok, Seoul, Taipei, and Tokyo in the days after its appearance. The system monitors the spread of infectious diseases by using NLP (Natural Language Processing) through which it analyzes the news coverage of several countries, in a different language while finding common patterns and building a joint distribution predicting how everything would unleash. 

Second Phase: Control

We are in this phase right now. AI plays a vital role here running through thousands of data points finding the definitive biological pattern where coronavirus is more likely to mutate. In fact, the virus mainly affects individuals with underlying diseases such as cancer, whose immune system is weak and vulnerable to new hosts. Similarly, after processing transportations details, it determined that any highly populated location and/or transportation method would be a transmitter for the disease and hence, avoidance is of utmost importance, especially in China.

Third Phase: Treatment

Overlapped with the previous phase, it comes the time when the disease begins to be treated. Doctors will use AI to find patterns between the existing virus and previous severe cases giving birth to a possible cure that needs to be tested. This may consist of treating a specific therapeutic target of the virus or treating its symptoms palliatively. The latter is what is usually done with coronaviruses, which do not have very effective treatments or vaccines and try to alleviate the symptoms to prevent them from becoming worse.

This phase works to mitigate infections and reduce the ability to expand based on treating and controlling those affected. Normally this phase occurs at the same time the disease begins to remit in a number of cases. At the moment, it might take a few months to get the vaccines ready and running after conducting the required clinical trials.

Although all phases are important, and they are also inseparable, we can probably highlight the special importance of control. The news of the closure of Wuhan and its surroundings, which has left some 40 million inhabitants in quarantine, has already jumped a few days ago into the arena. But, despite how tremendous it seems, this is actually one of the most effective actions.

First of all, we don’t know much about the transmission mechanisms. That means we cannot safely stick to a single-handed model: Is it transmitted during the incubation phase? Are there asymptomatic carriers? How many people can a patient infect? Without the required amount of data, only one answer remains maximum prevention. AI can support the search and help us save a large amount of time, however, no definitive vaccine can be commercially sold until the clinical trials have been successfully completed.