Prognosis Method on the Outcome of Covid-19 Patients in Senegal

Abstract

There have been disturbing waves of Covid-19 deaths in many countries due to a lack of adequate treatment in the early stages of the pandemic but also to the presence of co-morbidities in many hospitalised patients. This work aims to determine the discriminatory features between the surviving patients and the deceased ones after their hospitalisation to propose a new method of prognosis on the outcome of a new patient under treatment. To this end, we use three supervised classification methods, each allowing us to select the most significant features associated with patient death. These are binary logistic regression (BLR), random forests (RF), and support vector machines (SVM). The data comes from the Ministry of Health and Social Action of Senegal and covers the period from March 2020 to December 2022. Age emerged as the most discriminatory factor between the two patient groups: survivors and deceased. The study found that patients 60 and older are more likely to die of Covid-19.