Your browser doesn't support javascript.
Detection of COVID-19 from X-Ray Images using Deep Neural Networks
2021 International Conference on Computational Performance Evaluation, ComPE 2021 ; : 722-728, 2021.
Article in English | Scopus | ID: covidwho-1831750
ABSTRACT
In Dec 2019, Coronavirus has first shown in China and since then there is a big rise in the number of these cases. On March 28, 2020, WHO (World Health Organization) tweeted and proclaimed it's an epidemic. The test kits are relatively less likely to be checked by collecting blood samples because they are easily infectious, and collecting blood samples are very time taking. But it's important to get a quick and simpler way to validate the covid-19. Lungs are getting very badly affected by the Coronavirus and it increases in lungs gradually so we have to come up with the Convolutional Neural Network (CNN). This detects Corona virus utilising X-Rays of the Chest within about few seconds. To diagnose Coronavirus from X-ray Image dataset using different Convolutional Neural Network methodologies like Mobile Net, Inception, Exception, VGG. However, the findings obtained are based on the VGG16, VGG19 model. Apply the models to the X-ray dataset this was obtained from the Kaggle source. This dataset included 100 X-ray images of the lungs(chest) of the Patients with CORONA VIRUS, and 100 X-ray images of the lungs(chest) of People who are healthy. Python language is being used to execute the COVID19 dataset and Google Collaboratory is used for coding purposes. The focus of this research is to see how successful automatically detecting COVID-19 from chest X-rays using Convolutional Neural Networks. This study shows that to detecting COVID-19, VGG16 performs better than other method. The accuracy is 96.15% using VGG16 method. The excellent achievement of these models has the potential to rapidly better the COVID-19 diagnosis performance and speed. Although, A bigger dataset of chest X-ray pictures (COVID-19 positive) are necessary while using deep transfer learning to achieve consistent, accurate and better results to detecting COVID-19 diseases. © 2021 IEEE.
Keywords

Full text: Available Collection: Databases of international organizations Database: Scopus Language: English Journal: 2021 International Conference on Computational Performance Evaluation, ComPE 2021 Year: 2021 Document Type: Article

Similar

MEDLINE

...
LILACS

LIS


Full text: Available Collection: Databases of international organizations Database: Scopus Language: English Journal: 2021 International Conference on Computational Performance Evaluation, ComPE 2021 Year: 2021 Document Type: Article