ARCHIVES

Pneumonia remains a leading cause of mortality worldwide, particularly among children and elderly populations, necessitating rapid and accurate diagnostic methods. This study presents an advanced Convolutional Neural Network (CNN)-based framework for robust pneumonia detection and classification from chest X-ray images. The proposed model was trained and validated on the Kaggle Chest X-Ray Images dataset comprising 5,863 images from the Guangzhou Women and Children’s Medical Center, categorized into Normal and Pneumonia classes. Our custom CNN architecture incorporates multiple convolutional blocks with batch normalization, dropout regularization, and advanced optimization techniques to achieve superior performance. The model achieved an overall accuracy of 96.47%, precision of 95.82%, recall of 97.18%, F1-score of 96.49%, and specificity of 95.73% on the test dataset. Comparative analysis with state-of-the-art architectures including VGG-16, ResNet-50, InceptionV3, and DenseNet-121 demonstrates that our proposed framework outperforms existing models while requiring significantly less training time (45 minutes). The results indicate that deep learning-based automated pneumonia detection systems can provide reliable diagnostic support to radiologists, potentially reducing diagnostic errors and improving patient outcomes in clinical settings. This work contributes to the growing body of evidence supporting the integration of artificial intelligence in medical imaging for enhanced healthcare delivery.