Explainable Lung Disease Detection on Chest X-Ray Images Using Gradient-Weighted Class Activation Mapping

Early detection of respiratory diseases such as Coronavirus Disease-19 (COVID-19) and pneumonia are crucial for accelerating treatment and preventing more serious complications. This study proposes a method for classifying chest X-ray (CXR) images using a Convolutional Neural Network (CNN) to distinguish between COVID-19, pneumonia, and lung normal. Training model involves exploring various hyperparameter combinations to find the optimal configuration. The best results were achieved with a learning rate of 0.001, 50 epochs, and a batch size of 32, with an accuracy of 96.33%. Evaluation was performed using accuracy, precision, recall, and F1- score metrics, as well as a confusion matrix. This study used Gradient- Weighted Class Activation Mapping (Grad-CAM) as a transparent interpretation tool for model decisions. The main contribution of this study is the application of Grad-CAM in multi-class CXR classification to improve model interpretability in lung disease diagnosis