Using Digital Surveillance Tools for Near Real-Time Mapping of the Risk of International Infectious Disease Spread: Ebola as a Case Study

Sangeeta Bhatia; Britta Lassmann; Emily Cohn; Malwina Carrion; Moritz U.G. Kraemer; Mark Herringer; John Brownstein; Larry Madoff; Anne Cori; Pierre Nouvellet

The framework presented in this paper was developed as a proof-ofconcept to use digital surveillance data for near real-time forecasting of the spatio-temporal spread of an outbreak. It has been implemented as a web-based tool called "Mapping the Risk of International Infectious Disease Spread" (MRIIDS) (see [32] for more information). To further develop such approaches, it is important to establish an automated pipeline from data collection to curation to analysis, which currently requires manual intervention at each of these steps. Another factor that could enhance the usability of our model in near real time is to improve the running time of the fitting and forward simulation. In the current implementation, the running time varies from approximately 0.5 CPU hours when 100 days of incidence data are being used to approximately 335 CPU hours using 462 days of incidence data using a 3.3 GHz Intel Xeon X5680 processor. Although the West African Ebola epidemic was of unusual scale and duration, there is a scope for optimising the model implementation.