In a study published on April 13, 2021, researchers from the University of Gothenburg (UG) in Sweden explain why AI would be one of the solutions to be retained to contain the increase of infections during future epidemics or pandemics. Few studies have looked in the past at how AI could be used in the event of a pandemic. The health context due to the Covid-19 pandemic has obviously played a role in the increase of research on this subject.

The use of AI to counter pandemics finally addressed

Despite the fact that this study is only one of the first of its kind and that tests with real data are needed to improve the results, this research highlights the beneficial uses of AI to quickly control the epidemic.

In this study, researchers at the University of Gothenburg examined how machine learning algorithms could help find effective screening methods and thus better control future health crises despite the relatively limited data available during a pandemic.

Laura Natali, a researcher in the UG Physics department, spoke around the use of AI and the greater screening potential it could offer individuals:

“This can be a first step towards society having better control over future major epidemics and reducing the need for restrictions and shutdowns.”

A study that is a precursor to new initiatives

The new testing strategy based on machine learning will automatically adapt to the specific characteristics of sick individuals. It will be possible thanks to AI to know whether screening should be prioritized on an elderly population or not, or whether a specific geographical area -such as neighborhoods, public places- is considered at risk or not.

Laura Natali insists on the importance of identifying sick people:

“When an epidemic starts, it is important to identify infectious individuals quickly and efficiently. In random testing, there is a high risk of failing to do this, but with a more goal-oriented testing strategy we can find more infected individuals and thus also get the information needed to reduce the spread of the infection. We show that machine learning can be used to develop this type of testing strategy.”

In the Covid-19 pandemic, AI was used in several countries including South Africa, to alert the population about the risk of infection across the country’s provinces.

Translated from L’Université de Götenborg s’intéresse dans une étude à l’intelligence artificielle pour contenir les prochaines épidémies