Paper Title
Deep Learning Techniques for Pediatric Pneumonia Detection Using Radiographs

Abstract
Pneumonia is one of the primary causes of severe morbidity in children worldwide but, particularly in developing countries, with mainly limited and sometimes restricted access toa healthcare facilities. The aim of this paper was to assess the appropriateness of four deep models based on CNN: CNN, Resnet, Densenet, and VGG for the computerized diagnosis of pneumonia among children using chest X-rays. We applied it to a publicly available dataset in order to focus the pre processing method and enhance the clarity and contrast of the image. Additionally, we conducted a diagnostic accuracy for the evaluation of the model based on the model itself; among the used models, VGG architecture scored the highest rating, indicating a quite good chance for this tool regarding pneumonia diagnosis. Our work undoubtedly shows promise in deep learning for diagnostics with high accuracy and speed. The paper thus motivates that such deep learning technologies are put into vulnerable regions, enhancing the probability of early detection that requires treatment, which could then increase chances of survival. Future work would be applying the proposed models to practical clinical conditions that would allow possibly showing their applicability in the real world as well as comparatively feasible ways to place them into the healthcare infrastructure at the grass-root level.