Pneumonia Detection Using Anterior Chest X-Ray Images

Pneumonia is a common lung infection in developing and underdeveloped countries. Analysis of anterior chest x-ray is the most commonly used method to diagnose lung infections. Existing methods provide reasonable accuracies at the price of very high model sizes. Such a high memory requirement is contrary to the primary use case of such models, which is for a preliminary check from a place with access to less resources. We propose a lightweight neural network model to detect pneumonia in anterior chest x-ray images. The proposed model is trained to detect pneumonia in chest x-ray images utilizing contrast enhancement and progressive resizing.