A Comparative Study of Deep Convolutional Neural Network Models for Pneumonia Classification in Chest X-ray Images

Abstract

neumonia is a serious respiratory disease that remains one of the leading causes of morbidity and mortality worldwide. Early and accurate diagnosis is essential for effective treatment and improved patient outcomes. In recent years, advances in Artificial Intelligence and Deep Learning have demonstrated significant potential to assist radiologists through the automated analysis of medical images. This thesis presents a comparative study of three state-of-the-art Convolutional Neural Network (CNN) architectures, namely ResNet50, DenseNet121, and EfficientNet-B0, for pneumonia classification using chest X-ray images. Experimental results demonstrate that EfficientNet-B0 achieved the best overall performance with an accuracy of 96 %, outperforming ResNet50 and DenseNet121, which obtained accuracies of 93 % and 95 %, respectively. The obtained results confirm the effectiveness of deep transfer learning models for automated pneumonia detection and highlight the suitability of EfficientNet-B0 as a reliable Computer- Aided Diagnosis (CAD) tool for chest X-ray analysis.