The ability to rapidly recognize and respond to both global and local health threats remains a critical public health priority. The ever-growing digital world represents an unprecedented opportunity to harvest for tools to face emerging and re-emerging public health issues. Informal digital channels such as social networking sites, blogs, chat rooms, Web searches, local news media, crowdsourcing platforms have been credited with providing information that is not easily accessible by more traditional channels, such as census or traditional surveillance.
In particular, voluntarily provided self-reports on health-related matters can become a crucial tool in informing dynamical models for epidemic spreading on a national and global scale and can detect the temporal trends of influenza-like illness even without relying on a specific case definition. In this talk, we propose an unsupervised probabilistic framework that combines time series analysis of self-reported symptoms collected by the Influenzanet (http://influenzanet.info) platforms and performs an algorithmic detection of groups of symptoms, called syndromes.
This framework is capable of selecting temporal trends of syndromes that closely follow the ILI incidence rates reported by the traditional surveillance systems in the various countries without imposing any a priori ILI case definition. The proposed framework is able to forecast quite accurately the ILI trend of the forthcoming influenza season based only on the available information of the previous years. The result is a near-real-time flexible surveillance framework not constrained by any specific case definition and capable of capturing the heterogeneity in symptoms circulation during influenza epidemics in the various European countries.