(260f) (Invited Talk) COVID-19 Risk Emergency Toolbox - an Integrated System for Reliable Early Alert and Effective Public Health Risk Management

One of the key aspects of managing efficiently the public health risk of COVID-19 is the ability to assess the extent of the contagion spread in the community. In this paper we propose a combined approach integrating environmental surveillance data from a network of biosensors applied on sewage wastewater and fine/ultrafine air particles with human clinical data monitoring in potential patients or survivors of COVID-19 using a wearable device for telemedicine. These datasets are further fused with climatic data such as air temperature, humidity and UV radiation flux that may be linked to infectious diseases such as COVID-19. The integrated environmental/clinical dataset is then post-processed with machine learning algorithms to derive estimates of the degree of infection spread in the community at very early stages of infection, using the environmental parameters as sentinels for the presence of SARS-CoV-2 in the community. These estimates are then used as input to an already validated multi-stage SEIR-X model (CORE), which supports the management of the COVID-19 health risk by capturing reliably its dynamics, including the impact of risk reduction measures. We demonstrate the predictive capacity of the integrated CORE system in applications in Northern Italy and Greece even in the light of significant uncertainties with regard to the actual spread of the infection in the community due to limited molecular testing on humans. The integrated model is already part of the OECD COVID-19 computational toolbox.