Optimal multi-source forecasting of seasonal influenza

In the United States, seasonal influenza causes thousands of deaths and hundreds of thousands of hospitalizations. The annual timing and burden of the flu season varies considerably with the severity of the circulating viruses. Epidemic forecasting can inform early and effective countermeasures to limit the human toll of severe seasonal and pandemic influenza. With a growing toolkit of sophisticated statistical methods and the recent explosion of influenza-related data, we can now systematically match models to data to achieve timely and accurate warning as flu epidemics emerge, peak and subside. Here, we introduce a framework for identifying optimal combinations of data sources, and show that public health surveillance data and electronic health records collectively forecast seasonal influenza better than any single data source alone and better than influenza-related search engine and social media data.See it on Scoop.it, via Viruses, Immunology & Bioinformatics from Virology.uvic.ca
Optimal multi-source forecasting of seasonal influenza
Source: Viral Bioinformatics