Pneumonia Classification from Chest X-ray Images Using Pre-Trained Network Architectures

Abstract

Pneumonia is a serious disease caused by a lung infection that affects young and old people and approximately cause of 4 million deaths each year. Patients that are facing disorders such as weak immune systems, asthma, and babies all are at risk specifically if pneumonia is not detected at an early stage. An early diagnosis of pneumonia is required to plan a potential treatment strategy to control and treat the condition. The objective of this study is to analyze chest radiograph images to identify lung abnormalities using pretrained architecture. After extracting features from the images using convolutional neural network models that have been pre-trained on a large dataset called ImageNet, they are typically passed through a classifier for further processing and diagnosis. Pre-trained networks variants including VGG16, VGG19, DenseNet121, ResNet50, and InceptionV3 architecture were utilized in this study & results show that VGG-16 architecture performance is effective with a test accuracy of 90% and validation accuracy of 93.98% than other pretrained architectures.