Pneumonia Detection in Chest X-Ray Images Using Efficient and Lightweight Convolutional Neural Networks

Pneumonia remains a pressing global health challenge, demanding precise and prompt diagnosis for effective treatment. This study investigates the utilization of Convolutional Neural Networks (CNNs) to automate pneumonia classification by analysing chest X-ray images. Employing a diverse dataset of chest radiographs, our CNN model underwent comprehensive training, validation, and testing to differentiate between normal and pneumonia-affected cases. Preprocessing methods, such as standardization, augmentation, and normalization, were applied to bolster the model's resilience and adaptability. The model architecture featured multiple convolutional layers integrated with max-pooling, followed by dense layers employing rectified linear unit (ReLU) activation functions and dropout regularization to counteract overfitting. Following stringent evaluation, our proposed CNN model demonstrated impressive performance, achieving a test set accuracy of 98.89%. This work may aid healthcare professionals in prioritizing and expediting patient care. The outcomes underscore the importance of leveraging machine learning to enhance diagnostic capabilities for respiratory conditions, paving the way for more effective healthcare interventions.