Prediction of COVID 19- Chest Image Classification and Detection using RELM Classifier in Machine Learning
8th International Conference on Advanced Computing and Communication Systems, ICACCS 2022
; : 1184-1188, 2022.
Article
in English
| Scopus | ID: covidwho-1922640
ABSTRACT
This novel coronavirus (CoV) is known as 'SARS-CoV2' or '2019 novel coronavirus' or 'COVID-19' by the WHO. It is started at the end of 2019 in China. It is the outbreak of pneumonia related to chest issues. COVID-19 is an infectious virus. As COVID-19 is a contiguous disease, early detection is more important. It can be incurable if the virus is detected later. The identification of COVID-19 disease is done by collecting samples from the throat and nose. Sometimes when the patient is more severe, that time they are asked to take a chest X-Ray. This research proposed a system for the identification of the virus utilizing X-Ray images. Dataset used consists of both Covid and Normal X-ray images. In this research, we used the ResNet50 model to predict the disease. It contains 48 convolutional layers, one MaxPool, and Average Pool layers. 'RELM' is a suitable classifier, and it gave better accuracy than other classifiers. This research can be practically helpful to the physicians with the usage of datasets for the successful diagnosis of pandemic disease (COVID-19) in the healthcare field. We built the RELM classifiers with convolution Neural Network as our contribution in this research. © 2022 IEEE.
Chest image; Convolutional Neural Network; COVID-19; Machine Learning Algorhtms; RELM classifier; X-Ray; Convolution; Convolutional neural networks; Diagnosis; Image classification; Machine learning; Viruses; Coronaviruses; Image detection; Images classification; Infectious virus; Machine-learning; X-ray image
Full text:
Available
Collection:
Databases of international organizations
Database:
Scopus
Type of study:
Prognostic study
Language:
English
Journal:
8th International Conference on Advanced Computing and Communication Systems, ICACCS 2022
Year:
2022
Document Type:
Article
Similar
MEDLINE
...
LILACS
LIS