Comparing machine learning algorithms for predicting COVID-19 mortality

Abstract

Background The coronavirus disease (COVID-19) hospitalized patients are always at risk of death. Machine learning (ML) algorithms can be used as a potential solution for predicting mortality in COVID-19 hospitalized patients. So, our study aimed to compare several ML algorithms to predict the COVID-19 mortality using the patient’s data at the first time of admission and choose the best performing algorithm as a predictive tool for decision-making. Methods In this study, after feature selection, based on the confirmed predictors, information about 1500 eligible patients (1386 survivors and 144 deaths) obtained from the registry of Ayatollah Taleghani Hospital, Abadan city, Iran, was extracted. Afterwards, several ML algorithms were trained to predict COVID-19 mortality. Finally, to assess the models’ performance, the metrics derived from the confusion matrix were calculated. Results The study participants were 1500 patients; the number of men was found to be higher than that of women (836 vs. 664) and the median age was 57.25 years old (interquartile 18–100). After performing the feature selection, out of 38 features, dyspnea, ICU admission, and oxygen therapy were found as the top three predictors. Smoking, alanine aminotransferase, and platelet count were found to be the three lowest predictors of COVID-19 mortality. Experimental results demonstrated that random forest (RF) had better performance than other ML algorithms with accuracy, sensitivity, precision, specificity, and receiver operating characteristic (ROC) of 95.03%, 90.70%, 94.23%, 95.10%, and 99.02%, respectively. Conclusion It was found that ML enables a reasonable level of accuracy in predicting the COVID-19 mortality. Therefore, ML-based predictive models, particularly the RF algorithm, potentially facilitate identifying the patients who are at high risk of mortality and inform proper interventions by the clinicians.

Detection of COVID-19 Patients Using Machine Learning Techniques: A Nationwide Chilean Study

“…In general, our results showed that random forest showed regular performance compared to SVM and DF. In other studies, RF has shown the best performance in a study with smaller data volumes than those obtained in our research [21].…”

Explainable Artificial Intelligence Approach for the Early Prediction of Ventilator Support and Mortality in COVID-19 Patients

“…In addition, Khan et al [22] examined three comorbidities (cardiac problems, diabetes, hypertension) as significant features. However, Pezoulas et al [24] found that some lab tests are a significant attribute in predicting mortality, while Moulaei et al [25] discovered that shortness of breath and extra oxygen therapy are among the top features to predict mortality.…”

“…Recently, a study was made by Moulaei et al [25] to compare the performance of different ML techniques to predict mortality using data at the time of admission to hospital. The dataset contained the patients' clinical, demographic, and laboratory results.…”

“…The minimum patient age in the dataset was 19 and the maximum age was 107. A bin length of 10 was used, with the first bin range being [19][20][21][22][23][24][25][26][27][28][29], [29-39) and so on. The number of bins was 9.…”

Application of machine learning models based on decision trees in classifying the factors affecting mortality of COVID-19 patients in Hamadan, Iran

“…One of the methods of machine learning is interpretable decision trees, which are considered classification techniques in data mining [ 11 ]. One of the most popular usages of decision trees is to display the results as a simple decision tree algorithm that is easy to interpret for most researchers.…”