Comparative Analysis of Machine Learning Algorithms Using COVID-19 Chest X-ray Images and Dataset

Kumah, A.; Abuomar, O.

The COVID-19 pandemic, which first surfaced in Wuhan China, continues to have a devastating effect on the health and economies of affected countries, especially the United States. To contain the spread of the virus, it is imperative to have a system that can automatically identify the virus in people who show or do not show visible symptoms after being infected. This study aims at summarizing and comparing popular machine learning algorithms and determining which algorithm can best predict the presence of the virus in people by analyzing Chest X-ray images. While VGG16 and Support Vector Machines predicted the presence of the virus with the highest accuracy, all of the other studied techniques in this study i.e., decision tree, random forest, gradient boosting, and XGBoost also performed well and can prove useful not only in detecting asymptotic COVID-19 individuals, but also in eliminating the threatening virus surge we have experienced recently worldwide. © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.