Computer scientists have used a graph neural network to recognize flu cases as regionally interconnected clusters. This twist on flu forecasting lets the team’s algorithm spot infection patterns within and between regions, aiding decision-making by health officials.
Yue Ning, PhD, and colleagues at Stevens Institute of Technology in New Jersey say their AI technique produces an 11% boost in accuracy over that supplied by other contemporary systems.
On this strength, they claim, the method—which involves “capturing the interplay of space and time”—can predict flu outbreaks up to 15 weeks ahead of their arrival.
“Our model is also extremely transparent,” Ning says in a news release. “Where other AI forecasts use ‘black box’ algorithms, we’re able to explain why our system has made specific predictions, and how it thinks outbreaks in different locations are impacting one another.”
Ning and colleagues trained the algorithm on real-world state and regional data from the U.S. and Japan. They tested its predictions against historical flu data.
The novel technique has potential for predicting local and regional COVID outbreaks, the institute suggests.
Dave P. has worked in journalism, marketing and public relations for more than 30 years, frequently concentrating on hospitals, healthcare technology and Catholic communications. He has also specialized in fundraising communications, ghostwriting for CEOs of local, national and global charities, nonprofits and foundations.