Scientists Are Using AI to Predict Coronavirus’ Path

As the coronavirus continues to dominate headlines, scientists are pulling out all the stops to tackle it.

According to AsiaOne, scientists have deployed the latest computer modelling techniques to grapple with the virus, using them to predict everything from the potential number of cases to the eventual scale of its spread.

Scientists have been using computer models to predict viruses’ potential speed and spread for the past decades. Newer, adaptable models – those that take into account everything from human behavior trends to air traffic rates – are far more accurate.

And there are those that are assisted by artificial intelligence.

Arnaud Banos, research director at France’s Centre National de la Recherche Scientifique, said that AI-assisted models allow scientists to pick up on “weak signals” and turn them into crucial findings.

The weak signals could be as simple as “individual exchanges on social media discussing symptoms,” he said.

“The idea is to permanently collect mass amounts of data so that weak signals are automatically picked up and related to the evolution of the disease,” Banos added.

A British team at the London School of Hygiene and Tropical Medicine recently used these adaptable models to estimate the outbreak’s peak in Wuhan – the virus’ epicenter. With inputs constantly updated, they say the peak could come in mid-to-late February.

Rowland Kao, professor of Veterinary Epidemiology and Data Science at the University of Edinburgh, welcomed the results saying that “this is an analysis by an experienced and talented team,” but added that “as always the limitations of the available data will affect their predictions.”

Photo: Mike MacKenzie