A Comparative Analysis of Various Deep Learning Architectures for Chest Radiograph Classification

Anaccurate assessment of chest radiographs is of vital essence in radiology for the diagnosis of thoracic diseases. Recently the performance shown by deep learning models in various tasks pertaining to image analysis has been amazing. The purpose of this study is to find differences in learning outcomes between the four Machine Learning Models, namely AlexNet, ResNet-50, VGG-16, and MobileNet v2 to accurately determine the presence of one of five diseases based on the image of the chest radiographs that it is provided and compare the results of all to determine which is more effective. The primary focus of the study is to evaluate the effectiveness of these deep learning models in accurately classifying the radiograph into no finding or one of the five diseases it has been trained on.IJSRED-International Journal of Scientific Research and Engineering Developmen