Open AccessDetecting COVID 19 using Deep Learning
Author(s) -
Abhishek Uppula
Publication year - 2021
Publication title -
international journal for research in applied science and engineering technology
Language(s) - English
Resource type - Journals
ISSN - 2321-9653
DOI - 10.22214/ijraset.2021.37595
Subject(s) - covid-19 , residual neural network , deep learning , pneumonia , computer science , artificial intelligence , virology , medicine , disease , infectious disease (medical specialty) , outbreak
Corona virus disease 2019 (COVID 19) is defined as illness caused by novel corona virus now called severe acute respiratory syndrome corona virus 2 (SARS-Cov-2; formally called as 2019-nCov), which was first identified in Wuhan City, Hubei Province, China. The spreading of COVID 19 is very fast throughout the world. World economy as well as public health has been facing a devastating effect caused by COVID 19. Hence detecting COVID 19 is challenging task even we have multiple methods like RT-PCR, COVID kits. The RT-PCR may not available in all laboratory, even exists which take some time to process and get reports and COVID 19 test kits may not available in all places. So, the main intention of this paper is to detect COVID 19 with in low budget, less time and accurate results. We have trained deep transferred learning models like ResNet-50, ResNet-101 using COVID positive, Normal, Viral Pneumonia chest x-rays. ResNet-50, ResNet-101 is pre-trained deep learning neural network. ResNet-50 provides 98% of accuracy where ResNet-101 gives us 97% of accuracy. Keywords: COVID 19, Deep Learning, ResNet-50, Transferred Learning, Artificial Intelligence.