Skip to content

SadeemAlharthi/Benchmarking_Multiple_Deep_Learning_Models_for_Enhanced_COVID-19_Detection

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

21 Commits
 
 
 
 
 
 

Repository files navigation

Benchmarking_Multiple_Deep_Learning_Models_for_Enhanced_COVID-19_Detection

Due to the recent emergence of COVID-19, chaos and economic destruction have spread all around the world. As this is considered a novel virus, lack of resources and limitations of planning to cope with this pandemic have greatly impacted many lives and many people have lost their loved ones due to this situation. Thus, there is a need for some innovation of science and AI to handle this task on hand. It has become very necessary to investigate the use of the latest deep learning modeling techniques to discover efficient ways of detection of this novel virus. In this paper, we used publicly available datasets to apply some deep learning modeling techniques including VGG19, Xception, Resnet50, and GoogleNet in a transfer learning manner. Using such models eases the process of detection of the disease from chest X-rays to classify the patients based on their status: either COVID-19 or pneumonia chest/normal patients. The achieved preliminary results are promising. In particular, the Xception model achieves the highest performance with 99%, whereas the accuracy of the VGG19, ResNet50, and InceptionV3 are 98%, 98%, and 90% respectively.

Porject Folder Structure:

Porject Architectures :

  • Xception model.
  • VGG19 model.
  • Reset50 model.
  • GoogleNet or InceptionV3.

Citation

Our scientific paper has been included in the WHO in Global literature on Coronavirus disease. Please cite our paper: https://pesquisa.bvsalud.org/global-literature-on-novel-coronavirus-2019-ncov/resource/pt/covidwho-896143

Contact

If you have any suggestions or questions, please reach out to sadeem1.alharthi on Gmail

About

we used publicly available datasets to apply some deep learning modeling techniques

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published