Article’s

Pneumonia is a serious lung infection that can lead to severe health complications if not detected early. This project presents an AI-based pneumonia detection system using deep learning techniques and Grad-CAM visualization. The proposed model utilizes Convolutional Neural Networks (CNN) to classify chest X-ray images as normal or pneumonia affected. Transfer learning models such as ResNet50 and AlexNet are used to improve detection accuracy and reduce training time. Grad-CAM is integrated to highlight the infected regions in X-ray images, providing visual interpretability and helping medical professionals understand the prediction results. The system is developed with a Flask-based web application for easy user interaction. Experimental results demonstrate high accuracy and reliable performance in detecting pneumonia from chest X-rays. This approach can assist healthcare professionals in faster and more accurate diagnosis.