Patty Kostkova*, UCL, London, United Kingdom Track: Research Presentation Topic: Public (e-)health, population health technologies, surveillance Presentation Type: Oral presentation Submission Type: Single Presentation Building: Sol Principe Room: C - Almudaina Date: 2014-10-09 11:00 AM – 11:45 AM Last modified: 2014-09-03

If you are the presenter of this abstract (or if you cite this abstract in a talk or on a poster), please show the QR code in your slide or poster (QR code contains this URL).

Background
Recently, early warning, prevention, management and control of infectious diseases have been brought to the forefront of public health activities. In addition to traditional syndromic surveillance systems, public health Big data play an increasingly important role in preparedness, early-detection, response and risk communication. Real-time data streams from new situation-aware realtime signals from social media, online searches, loyalty cards, mobile/sensor networks and citizens’ participatory surveillance systems need more effectively complement traditional surveillance to enhance preparedness and ensure more effective response.

Objective:
In this abstract, we present medi+board – a public health dashboard system screening real-time data sources for early detection of infection threats, cross-validating sources by correlating data streams and displaying results in an integrated format presented by means of an interactive dashboard.

Methods
An integrated digital public health infrastructure was designed for multiple data-streams with dynamic components for data-mining, threat detection, verification, correlation of threats. Monitoring and detection of multiple channels requires different computational methods according to their structure, timelessness and reliability . Our algorithm is expressed in the form of a directed acyclic graph which can be programmed by the analyst in a visual designer to prevent the need of writing code. Events verified by the processing system are forwarded to the dashboard and, according to the event template, presented in the appropriate visual format. Apart from the resulting increase in productivity, every node encapsulates a specific step in the algorithm with its own distinct set of properties such as the reliability coefficient that can be manipulated directly in the designer.

Results:
medi+board functionality was shown on a simulation of the swine flu pandemics in 2009. We used three datasets: the HPA surveillance data from the Royal College of General Practice (RCGP), Google News API, and Twitter dataset collected during the pandemics in 2009.
The dashboard illustrated the three data streams in a dynamic way while statically illustrating other important segments of data, as if these were available, at two key points of the pandemics (containment phase in the UK when the demographics study of initial cases ‘FF100’ was conducted; and control phase in the autumn monitoring the distribution of the anti-virals as well as calls to the dedicated “fluline”). Further new data-stream extensions, in progress, include integration of the Medisys new media stream and the InfluenzaNet - the successful pan- European participatory surveillance network for ILI.

Conclusions:
In this abstract, we presented a novel PH dashboard system medi+board. Demonstrated on three data streams from the swine flu 2009 pandemic (RCGP surveillance in the UK, Google News and Twitter streams), the medi+board integrated public health dashboard provides a simulation illustrating how such system could substantially enhance future public health operations. At Medicine 2.0, we will present our new data-stream extensions, in progress, such as the Medisys new media epidemic intelligence stream and the EU participatory network Influenzanet.