Advanced Feature Extraction and Transformation Method for Pneumonia Detection in X-Ray Images

Pneumonia is a serious lung inflammation caused by pathogens or autoimmune conditions, affecting about 450 million people worldwide each year. Chest X-rays are the primary diagnostic tool, but manual analysis can be time-consuming and prone to errors. With advances in deep learning, automated systems are increasingly used to support medical image analysis. This study explores the use of MobileNetV2, a lightweight convolutional neural network, for detecting and classifying pneumonia from chest X-ray images. MobileNetV2 uses depth wise separable convolutions and an inverted residual structure to achieve efficient computation with low memory usage while maintaining high accuracy. The proposed framework distinguishes between pneumonia and normal cases, making it suitable for real-time and resource-limited healthcare environments, providing faster, reliable, and cost-effective diagnostic assistance to medical professionals.