Study on Convolutional Neural Network to Detect COVID-19 from Chest X-Rays

The world is facing a pandemic due to the coronavirus disease 2019 (COVID-19), named as per World Health Organization. COVID-19 caused by virus called severe acute respiratory syndrome 2 (SARS-CoV-2), which was initially discovered in late December Wuhan, China. Later, had spread throughout within few months. has become global health crisis because millions of people worldwide are affected this fatal virus. Fever, dry cough, and gastrointestinal problems most common signs COVID-19. highly contagious, can easily those with whom they have close contact. Thus, contact tracing suitable solution prevent from spreading. method identifying all persons COVID-19-affected patient come into last weeks tracing. This study presents an investigation convolutional neural network (CNN), makes test faster more reliable, detect chest X-ray (CXR) images. Because there many studies field, designed model focuses on increasing accuracy level uses transfer learning approach custom model. Pretrained deep CNN models, such VGG16, InceptionV3, MobileNetV2, ResNet50, been used for feature extraction. performance measurement based classification accuracy. results indicate that recognize SARS-CoV-2 CXR provided 93% 98% validation accuracy, pretrained customized models MobileNetV2 obtained 97% InceptionV3 98%, VGG16 respectively. Among these recorded highest