Measuring Icebergs: Using Different Methods to Estimate the Number of COVID-19 Cases in Portugal and Spain


By Carlos Baquero, University of Minho and INESC TEC

During the current coronavirus pandemic, monitoring the evolution of COVID-19 cases is very important for the authorities to make informed policy decisions, and for the general public that has the right to be informed of the reach of the problem. Official numbers of confirmed cases are periodically issued by each country’s health authority. Unfortunately, upon the pandemic outbreak is it usually the case that there is lack of available laboratory tests, and other material and human resources. Hence, it is not possible to test all potential cases, and some eligibility criteria is applied to decide who is tested. Under these circumstances, the evolution of official confirmed cases might not represent the total number of cases. We present a new approach to estimate the number of cases with COVID-19 symptoms, which is based on using crowdsourcing with open anonymous surveys to obtain indirect information. We compare this new approach with approaches that infer current cases from the case fatality series for two particular countries, Portugal and Spain.

About the Speaker:
I am a Professor in the Informatics Department, Universidade do Minho, and a Researcher at the High Assurance Laboratory (HASLab) within INESC TEC. My research interests cover data management in eventual consistent settings, distributed data aggregation and causality tracking. In the last years I have collaborated with my co-authors in the development of data summary mechanisms such as Scalable Bloom Filters, causality tracking for dynamic settings with Interval Tree Clocks and Dotted Version Vectors and in predictable eventual consistency with Conflict-Free Replicated Data Types. My recent work has been applied in the Riak distributed database, Redis CRDBs, Akka distributed data, Microsoft Azure Cosmos DB, and is running in production systems worldwide.

More Details

Digital Platform:  Zoom