Attention-Enhanced CBAM-VGG-16 for Optimized Pneumonia Detection in Chest X-Ray Imaging

Pneumonia remains a significant global health concern, particularly in pediatric populations, necessitating accurate and timely detection from chest X-rays for effective intervention. Traditional diagnostic methods and conventional machine learning models often struggle with accuracy and the complex nature of X-ray images. To address these challenges, this work introduces a novel attention module integrated with the VGG-16 architecture for enhanced pneumonia detection in chest Xrays. The performance of this attention-augmented VGG-16 model is compared against five other convolutional neural network (CNN) architectures designed for pneumonia detection. Our results demonstrate a significant improvement in diagnostic accuracy, with the attention-based VGG-16 outperforming other models by an average of 3.78% in accuracy and 2.61% in F1-score. The attention module enables the model to focus on the most relevant regions of the X-ray images, leading to better localization and identification of pneumonia-affected areas. These findings highlight the potential of attention mechanisms in improving CNNbased diagnostic tools, advancing clinical decision-making, and supporting early diagnosis. This research contributes to the development of robust, scalable models that could ultimately enhance treatment outcomes and reduce the burden of pneumonia worldwide.