On Artificial Intelligence to Model the COVID-19 Country Infection Trends

Methods: From the COVID-19 dataset by the Johns Hopkins University Center for Systems Science and Engineering (JHU CSSE), we present in this manuscript a temporal analysis on the number of new cases and deaths per million of inhabitants in the most affected countries using Artificial Intelligence. Our approach incrementally models the COVID-19 cases using a hierarchical clustering algorithm that points out country transitions from/to infection groups over time. From this model, one can label and compare the current situation of a country against others that have already faced previous waves.

Findings: Given the current status of the pandemic around the world, we provide some special analysis on the most affected countries and identify significant changes over time. By using our clustering approach, we design a transition index to estimate the most probable countries' movements in between infectious groups thus allowing to point out next wave trends.

Interpretation: We draw two important conclusions on the temporal number of deaths and confirmed cases per million of inhabitants. Firstly, we show the historical infection path taken by specific countries thus emphasizing changing points, i.e., when countries moved between clusters with small/medium/large number of cases. Secondly, we estimate new waves for specific countries using the transition index.

Funding: This work was supported by CAPES (Coordination for the Improvement of Higher Education Personnel -- Brazilian Federal Government Agency) grant number 88887.463387/2019-00, CNPq (Brazilian National Council for Scientific and Technological Development) grant number 302077/2017-0, and FAPESP (São Paulo Research Foundation) grant number 2013/07375-0.