Predicting Mortality in COVID-19 Patients Using 6 Machine Learning Algorithms.

In late 2019, COVID-19 appeared and has since spread worldwide as the new pandemic, causing more than 6 million deaths. In dealing with this global crisis, the contribution of Artificial Intelligence was also important through the possibilities of creating predictive models through Machine Learning algorithms, which are already successfully applied to solving a multitude of problems, for many scientific fields. This work aims to find the best model for predicting the mortality of patients with COVID-19, through the comparison of 6 classification algorithms, i.e. Logistic Regression, Decision Trees, Random Forest, eXtreme Gradient Boosting, Multi-Layer Perceptrons, K-Nearest Neighbors. We used a dataset containing more than 12 million cases which was cleansed, modified, and tested for each model. The best model is XGBoost (Precision: 0.93764, Recall: 0.95472, F1-score: 0.9113, AUC_ROC: 0.97855 and Runtime: 6.67306 sec), which is recommended for the prediction and priority treatment of patients with high mortality risk.