Company or Institution
ELLIS unit Alicante Foundation, Universitat d'Alacant, and Universitat Politècnica de València
Sustainable Development Goals (SDGs)
SDG 3: Good Health and Well-being
SDG 11: Sustainable Cities and Communities
SDG 16: Peace and Justice Strong Institutions
General description of the AI solution
Non-pharmaceutical interventions (NPIs) were applied by the governments during the COVID-19 pandemic in an effort to contain the spread of the SARS-CoV-2 virus. These interventions include restrictions on gatherings, limitations to human mobility, closings of schools and education centers, facial mask mandates, labor restrictions, etc…
The efficacy and individual contribution of each of these NPIs to the spread of the virus are yet to be clearly modeled and understood. Thus, computational epidemiological models that were designed to predict the number of COVID-19 cases in the future fail to provide accurate predictions. Moreover, it is hard to estimate the real economic and social impact that would result from the application of these measures.
To help governments and public institutions make more informed, accurate decisions in the context of pandemics, we are working on building a pandemic decision-support dashboard. The dashboard consists of two main components: a predictor of COVID-19 cases and a prescriptor of non-pharmaceutical intervention plans. The predictor provides accurate predictions of COVID-19 cases in the world, taking into account the NPIs applied in each country/region and the historic time series of cases. Based on these predictions, the prescriptor recommends up to 10 different non-pharmaceutical intervention plans which have the best trade-off between the economic and social impact of applying such interventions vs the resulting number of COVID-19 cases if the interventions were to be applied. Therefore, these intervention plans are on the Pareto front of the two-dimensional space of healthcare cost (COVID-19 cases) vs economic/social cost.
Github, open data repository, prototype or working demo
a) . M.A. Lozano, O. Garibo i Orts, E. Piñol, M. Rebollo, K. Polotskaya, M.A. García-March, J.A. Conejero, F. Escolano, and N. Oliver. Open data science to fight COVID-19: Winning the 500k XPRIZE Pandemic Response Challenge. Proceedings of the Thirty-First International Joint Conference on Artificial Intelligence (IJCAI). Sister Conferences Best Papers. Pages 5304-5308. https://doi.org/10.24963/ijcai.2022/740
b). BEST PAPER AWARD in Applied Data Science at the Joint European Conference on Machine Learning and Knowledge Discovery in Databases ECML-PKKD 2021
M.A. Lozano, O. Garibo i Orts, E. Piñol, M. Rebollo, K. Polotskaya, M.A. García-March, J.A. Conejero, F. Escolano, and N. Oliver. Open data science to fight COVID-19: Winning the 500k XPRIZE Pandemic Response Challenge. In Joint European Conference on Machine Learning and Knowledge Discovery in Databases ECML-PKKD 2021 (pp. 384-399). Springer, Cham. https://doi.org/10.1007/978-3-030-86514-6_24
c) How Valencia crushed Covid with AI (By leveraging algorithms and unorthodox data sources, an MIT researcher has made Valencia a Covid-19 data pioneer) WIRED, Sept. 8Th, 2021 https://www.wired.co.uk/article/valencia-ai-covid-data
d). VALENCIA IA4COVID, Winners of the 500k Pandemic Response Challenge organized by the XPRIZE Foundation https://www.xprize.org/challenge/pandemicresponse/winners-results