<![CDATA[Pneumonia is a pulmonary infection that remains one of the leading causes of death among children under five, especially in developing countries. Early detection and rapid diagnosis are critical in managing this disease, particularly in regions with limited access to medical professionals. This study aims to develop an automatic classification system for pediatric chest X-ray images using the Convolutional Neural Network (CNN) method to detect pneumonia. The dataset used consists of 5,863 pediatric chest X-ray images categorized into two classes: Pneumonia and Normal. The images underwent preprocessing stages including resizing, normalization, augmentation, and noise removal. The CNN architecture includes stacked convolutional layers, max pooling, dropout, and a fully connected layer with sigmoid activation. The model was trained using 80% of the data for training, 10% for validation, and 10% for testing. Performance was evaluated using accuracy, precision, recall, and F1-score metrics. Evaluation results showed that the model achieved over 93% accuracy, with 92.5% precision, 94.2% recall, and an F1-score of 93.3%. Transfer learning using pretrained models (VGG16 and ResNet50) further improved performance. These findings demonstrate that CNN is an effective tool for medical image classification and has strong potential to support fast and accurate pneumonia diagnosis, especially in resource-limited healthcare settings.]]>
Copyrights © 2025
