Published Paper Details:

Pneumonia is an infection that causes inflammation in one or both of the lungs. It causes the air sacs, or alveoli, of the lungs to fill up with fluid or pus. Pneumonia poses significant health risks, notably in India, where it contributes to 23% of the global burden with high mortality rates. As the diagnosis of pneumonia depends on skilled radiologists reading chest x-ray, delay in diagnosis, particularly in distant places, impede prompt treatment. The effectiveness of deep learning, in particular Convolutional Neural Networks (CNNs), in automating pneumonia detection is examined in this article. Through transfer learning we investigate pre-trained models such as VGG16, ResNet-50V2, InceptionV3, MobileNetV2, and Dense Net as feature extractors to improve the performance of the CNN architectures. Through experimentation and training on Chest X-ray dataset, the optimal CNN model for pneumonia diagnosis can be found by accurately classifying lung X-rays as normal or pneumonia infected .