Diagnosis of COVID-19 cases on X-Ray images using CNN

Abstract

COVID-19 is a viral disease that has killed more than 10 million people worldwide and infected millions of people. Therefore, it has become necessary to screen large numbers of people to detect infected individuals and reduce the spread of the disease. Maximum spread for confirming a virus is estimated with RT-PCR test. PCR (Polymerize Chain Response) is a popular device for predicting pathological examination. A key issue with real-time RT-PCR testing is the risk of generating false-negative and false-positive results. As an adjunct to RT-PCR, Computed Tomography (CT) can be used to diagnose COVID-19. In this article, using CXR scans, we proposed a deep-layered convolutional neural network (CNN) for accurate COVID-19 detection. Our model yields 97% accuracy.

X-rays imaging analysis for early diagnosis of thoracic disorders using capsule neural network: a deep learning approach

“…CapsNN is a network that can extract geographical information as well as other critical attributes, preventing the information loss witnessed during pooling operations. Capsule gives us a vector with a direction as an output [45]. To preserve the locations of items and their attributes inside the image as well as simulate their hierarchical relationships, capsule networks have been created.…”