<![CDATA[The rapid spread of Coronavirus Disease 2019 (COVID-19) has created an urgent need for accurate and rapid diagnostic methods to support clinical decision-making. Chest X-ray imaging has become one of the most widely used imaging modalities due to its availability, low cost, and ease of acquisition. However, manual interpretation of chest X-ray images is time-consuming and highly dependent on radiologists' expertise. This study aims to compare the performance of three supervised machine learning algorithms, namely Random Forest (RF), K-Nearest Neighbor (KNN), and Support Vector Machine (SVM), for classifying COVID-19 and normal chest X-ray images. Feature extraction was performed using the SqueezeNet image embedding method, which generated 1,000-dimensional numerical feature vectors. The dataset consisted of 288 chest X-ray images (144 COVID-19 and 144 normal), divided into 200 training images and 88 independent testing images. Model performance was evaluated using 5-fold cross-validation on the training set and an independent test set based on Area Under the Curve (AUC), Classification Accuracy (CA), Precision, Recall, F1-score, and Matthews Correlation Coefficient (MCC). The experimental results showed that all algorithms achieved an AUC above 98% and a CA above 94%. Among them, the SVM with a Radial Basis Function (RBF) kernel achieved the best performance, obtaining an AUC of 98.9%, CA of 97.0%, Precision of 97.1%, Recall of 97.5%, F1-score of 97.0%, and an MCC of 0.941. These findings demonstrate that the combination of SqueezeNet image embedding and SVM is an effective approach for chest X-ray image classification and has the potential to support the development of clinical decision support systems for COVID-19 diagnosis.]]>
