A Comparative Study of the Results of the Diagnosis of Covid-19 Disease in Chest x-ray Using Some Models of Convolutional Neural Network

Authors

  • Mohammad Blailo Tishreen University

Keywords:

Deep learning (DL), computer vision, convolutional neural network (CNN), corona virus, artificial intelligence (AI), Representation Learning, Feature Learning, Machine Learning(ML).

Abstract

 

Economic literature

The research aims to provide a new framework for the concept of deep learning and its importance in providing various techniques that provide assistance to radiologists in diagnosing infection with the Corona virus (Covid-19) in simple chest radiographs from one of the deep learning techniques used in research, the convolutional neural network Resnet50, which is one of the types of neural networks and comparison of results with research results on the same topic conducted on other artificial neural networks.

The x-rays were obtained from the database of D. Joseph Paul Cohen, Professor at the University of Montreal, which is available on the World Wide Web on githup free of charge and always updated with new images. It includes a large number of simple radiographs and computed tomography of the chest with different dimensions of normal people and patients who mainly suffer from various respiratory syndromes, including the Corona virus.

The practical steps in the research were studied through the DeepLearing toolbox in MATLAB 2018B on a laptop with a 2 GHz dual-core processor and 4 GB memory, the results showed 98.1% accuracy and 100% sensitivity

Author Biography

Mohammad Blailo, Tishreen University

Work Supervisor, Department of Computer and Automatic Control, Faculty  of  Mechanical and Electrical Engineering

Published

2021-09-14

How to Cite

1.
بليلو م. A Comparative Study of the Results of the Diagnosis of Covid-19 Disease in Chest x-ray Using Some Models of Convolutional Neural Network. Tuj-eng [Internet]. 2021Sep.14 [cited 2024Feb.21];43(4). Available from: https://journal.tishreen.edu.sy/index.php/engscnc/article/view/10830