Support for Medical Diagnosis of COVID-19 Using Machine Learning Algorithms on Chest CT Images: A Comparative Study of Supervised Classifiers

COVID-19, caused by the SARS-CoV-2 virus, is a highly contagious respiratory disease with serious consequences. Accurate and rapid diagnosis is crucial for disease control, but nonspecific symptoms and delays in testing hinder the process. Machine learning algorithms have shown promise in providing more efficient and faster diagnoses. In this study, four supervised classifiers were compared using computed tomography (CT) images from 35 patients with COVID-19. Feature extraction was performed using the gray-level co-occurrence matrix technique, and feature selection was carried out using Random Forest and Extra-Trees classifiers. The models were trained, evaluated, and adjusted to achieve best performance. The study aims to support medical decision-making by utilizing machine learning algorithms in the diagnosis of COVID-19 based on CT images.