Harnessing multiple data streams and artificial intelligence to better predict flu - (ScienceDaily via NewsPoints Desk)

  • Study data published in Nature Communications suggest an approach that includes the combination of two forecasting methods with machine learning, called ARGONet, can estimate local influenza activity, ScienceDaily reported Friday.

  • When applied to influenza seasons from September 2014 to May 2017, ARGONet made more accurate predictions than the researchers' earlier high-performing forecasting model, ARGO, in more than 75 percent of the US states studied.

  • "Timely and reliable methodologies for tracking influenza activity across locations can help public health officials mitigate epidemic outbreaks and may improve communication with the public to raise awareness of potential risks," said senior author Mauricio Santillana.

  • According to the news source, ARGONet uses machine learning and ARGO, which leverages information from electronic health records, influenza -related Google searches and historical influenza activity in a given location, along with a second model, which draws on spatial-temporal patterns of influenza spread in neighbouring areas.

  • The machine learning system was trained by feeding it influenza predictions from both models as well as actual influenza data, helping to reduce prediction errors.

  • "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," said first author Fred Lu.

Did you like this article?