eduzhai > Applied Sciences > Engineering >

Stacked Convolutional Neural Network for Diagnosis of COVID-19 Disease from X-ray Images

  • Save

... pages left unread,continue reading

Document pages: 26 pages

Abstract: Automatic and rapid screening of COVID-19 from the chest X-ray images hasbecome an urgent need in this pandemic situation of SARS-CoV-2 worldwide in2020. However, accurate and reliable screening of patients is a massivechallenge due to the discrepancy between COVID-19 and other viral pneumonia inX-ray images. In this paper, we design a new stacked convolutional neuralnetwork model for the automatic diagnosis of COVID-19 disease from the chestX-ray images. We obtain different sub-models from the VGG19 and developed a30-layered CNN model (named as CovNet30) during the training, and obtainedsub-models are stacked together using logistic regression. The proposed CNNmodel combines the discriminating power of the different CNN`s sub-models andclassifies chest X-ray images into COVID-19, Normal, and Pneumonia classes. Inaddition, we generate X-ray images dataset referred to as COVID19CXr, whichincludes 2764 chest x-ray images of 1768 patients from the three publiclyavailable data repositories. The proposed stacked CNN achieves an accuracy of92.74 , the sensitivity of 93.33 , PPV of 92.13 , and F1-score of 0.93 for theclassification of X-ray images. Our proposed approach shows its superiorityover the existing methods for the diagnosis of the COVID-19 from the X-rayimages.

Please select stars to rate!

         

0 comments Sign in to leave a comment.

    Data loading, please wait...
×