Machine Learning Performances for Covid-19 Images Classification based Histogram of Oriented Gradients Features

Abstract

Coronavirus disease (Covid-19) is an infectious disease that attacks the respiratory area caused by the severe acute respiratory syndrome (SARS-CoV-2) virus. According to the World Health Organization (WHO) as of April 2022, there were more than 500 million cases of Covid-19, and 6 million of them died. One of the tools to detect Covid-19 disease is using X-ray images. Digital X-ray images implementation can be developed classification method using machine learning. By using machine learning, the diagnosis of this disease can be faster. This study applied a features extraction method using the Histogram of Oriented Gradients (HOG) algorithm and the Linear Support Vector Machine (SVM), K-Nearest Neighbor (KNN) Medium and Decision Tree (DT) Coarse Tree classification methods. The study can be used in the diagnosis of Covid-19 disease. The best method among the classification methods is features extraction from HOG algorithm and DT Coarse Tree. The highest values of accuracy, precision, recall, specificity, and F-score were 83.67%, 96.30%, 78.79%, 98.25, and 76.48%.