Classification and detection of covid-19 in human respiratory lungs using convolutional neural network architectures
2021 International Conference on Artificial Intelligence for Cyber Security Systems and Privacy, AI-CSP 2021
; 2021.
Article
in English
| Scopus | ID: covidwho-1713966
ABSTRACT
The covid-19 was discovered for the first time in the Wuhan region of China. It was on December 31, 2019, that it all started. The speed of propagation of this germ has grown exponentially and on a planetary scale. This disease caused panic among specialists because the virus caused a high fever with cough in the patient and loss of consciousness. The worst part of all of this was that the coronavirus was virulent and contagious. The doctors did not know how to fight this new disease. Researchers in medical imaging opted for the tool of artificial intelligence, more particularly the technology of deep learning with its architectures of convolutional neural networks. The goal is the classification and detection of coronavirus disease. In our paper, we propose to study coronavirus disease. First, we will classify x-ray pictures into three main classes (viral pneumonia, normal, and covid-19). Then, we will use deep learning technology with convolutional neural network architectures. The architecture used is VGG, InceptionV3, DenseNet, MobileNet, ResNet, and Xception. The test results are satisfactory, with an accuracy of 95.83% for MobileNetV1 and 98.95% for ResNet101. © 2021 IEEE.
Classification; CNN; Coronavirus; Covid-19; Deep Learning; Detection; Backpropagation; Convolution; Convolutional neural networks; Disease control; Medical imaging; Network architecture; Technology transfer; Convolutional neural network; Coronaviruses; Loss of consciousness; Neural network architecture; Planetary scale; Speed of propagation
Full text:
Available
Collection:
Databases of international organizations
Database:
Scopus
Language:
English
Journal:
2021 International Conference on Artificial Intelligence for Cyber Security Systems and Privacy, AI-CSP 2021
Year:
2021
Document Type:
Article
Similar
MEDLINE
...
LILACS
LIS