PNEUMONIA DETECTION USING CHEST RADIO GRAPHS WITH DEEP LEARNING MODELS

Abstract

Abstract— Pneumonia remains a critical public health concern, particularly in low-resource settings where early and accurate diagnosis is essential for effective treatment and prevention of severe complications. With the rise of deep learning in medical imaging, automated detection of pneumonia from chest X-rays has gained significant momentum. This study explores and evaluates the performance of three state-of-the-art convolutional neural network architectures—MobileNetV2, DenseNet121, and ConvNeXt—for the classification of chest X-ray images into pneumonia and normal categories. MobileNetV2 offers a lightweight solution optimized for real-time applications, making it ideal for deployment in clinical environments with limited computational resources. DenseNet121, known for its dense connectivity and feature reuse, enables deep supervision and robust learning from limited data. ConvNeXt, a modernized convolutional architecture