Classification and Analysis of Patients with COVID-19 Using Machine Learning

Abstract

The rapid spread of the Coronavirus disease (COVID-19) has demanded studies and research works from many areas of knowledge, searching for treatments, vaccines and preventive measures. This pandemic has become a very challenging situation due to its substantial demand for medical infrastructure. In this context, this paper proposes to apply Machine Learning methods to classify and to analyse the outcome of patients with COVID-19 as discharge or death and to describe the profile of patients infected by the coronavirus. The dataset consists of clinical data from Sírio Libanês Hospital, available in the FAPESP repository (2020). Results indicate that, among all tested classifiers, the Naive Bayes algorithm presents better performance and it better represents the phenomenon under study, demonstrating superiority in terms of classification and induction numerical analysis of the epidemiological phenomenon for COVID-19.