Organization: Los Alamos National Laboratory
Year: 2017

When many people feel ill and experience new or unexpected symptoms, they search websites such as Wikipedia. WikiEpiCast combines that Wikipedia readership traffic with mathematical models and clinical surveillance data to forecast the spread of diseases around the world. The WikiEpiCast framework provide near-real-time nowcasts for some diseases and forecasts for other diseases around the globe. Preliminary results of influenza forecasts in the U.S. using the WikiEpiCast approach have been on target. For example, WikiEpiCast was used to predict the timing of the 2015-2016 influenza season. The WikiEpiCast prediction that the flu would not peak until February and that it would be mild was later confirmed by statistics from the Centers for Disease Control and Prevention. Using the WikiEpiCast framework makes it possible to provide accurate estimates on when the flu season will start, when it will peak and just how severe it will be at national (and at this point ten regional) levels within the United States. This could possibly prevent a full-fledged epidemic by giving decision makers much needed time to launch broad-based vaccination programs and implement a widespread communication strategy.