Hershey, Pennsylvania
New York, New YorkBeijing, China
Robust Diagnostic System for COVID-19 Based on Chest Radiology Images

Robust Diagnostic System for COVID-19 Based on Chest Radiology Images

Sasikaladevi N. (School of Computing, SASTRA University (Deemed), India) and Revathi A. (School of EEE, SASTRA University (Deemed), India)
Copyright: © 2022 |Pages: 16
DOI: 10.4018/978-1-7998-9012-6.ch003
OnDemand:
(Individual Chapters)
Available
$37.50
Current Special Offers
No Current Special Offers
TOTAL SAVINGS: $37.50

Abstract

The proposed system is based on a diagnosis of COVID from x-ray images. In the respiratory system, 17 different viral infections are possible. Accurately discriminating COVID from other viral infections is necessary today as it spreads rapidly. The proposed system differentiates COVID infection accurately from other viral infections. The convolutional neural network (CNN) provides superior performance for disease diagnosis based on images in the deep learning era. In this chapter, to solve this issue, the authors propose a hypergraph-based convolutional neural network-based fast and accurate diagnosis system for COVID. In this work, the hypergraph represents the sophisticated features of a lung x-ray image to diagnose COVID. In-depth features are extracted from the x-ray images using residual neural networks. In order to discriminate COVID viral infection from other viral infections, the hypergraph fusion approach is used.
Chapter Preview
Top

This section analyzes various state-of-the-art methods to detect COVID from X-ray images. These methods can be classified into four different categories. Some researchers used the pre-trained neural network in X-ray images for classification. The second category is transfer learning, in which the pre-trained neural network is customized and then used for classification. Few researchers proposed deep learning and machine learning methods to predict COVID from the x-ray image dataset by designing the network from scratch. Finally, very few research works are based on extracting the in-depth features from the X-ray images, and then these extracted features are used for classification by using machine learning or deep learning models.

Complete Chapter List

Search this Book:
Reset