Researchers from the Computational Health Informatics Program (CHIP) at Boston Children’s Hospital combined forecasting methods with artificial intelligence (AI) to predict flu activity. Results from the research study were published in Nature Communications.
The Approach is called ARGONet and was applied to flu seasons from September 2014 to May 2017. Results show that it made more accurate predictions than the team's earlier high-performing forecasting approach, ARGO, in more than 75% of the states studied.
The ARGONet solution is also able to use information from electronic health records, flu-related Google searches, and historical flu activity in a given location. Here is how it works. ARGONet was "trained" by feeding it flu predictions from both models as well as actual flu data, helping to reduce errors in the predictions. The AI solution evaluates the predictive power of each independent method and recalibrates how this information should be used to produce improved flu estimates.
Researchers said ARGONet produces the most accurate estimates of influenza activity available to date, a week ahead of traditional healthcare-based reports, at the state level across the U.S.
"We think our models will become more accurate over time as more online search volumes are collected and as more healthcare providers incorporate cloud-based electronic health records," Fred Lu, a CHIP investigator and first author on the paper, said in a release.