|
چکیده
|
— The quick and early detection of COVID-19 infection is very important in the fight against the pandemic. Deep learning can be considered as a helpful method to provide help and assist the medical staff to detect the infection of COVID-19, which will definitely have a positive effect in controlling the outbreak of COVID-19. In this paper, we will propose a simple CNN based deep learning model called Grad-CAM CNN (GCNN), for the purpose of detecting the infection of COVID-19 disease from chest X-ray images and visualizing a heat map with the help of the GradCAM technique in order to determine which area in the X-ray image of the chest has COVID-19. Since CNN is a very powerful method in processing images, we use it to build the model. We evaluated the performance of the proposed method on public online available datasets of X-ray images of the chest from Kaggle. The proposed method is able to achieve an accuracy score of 97.78 % using a learning rate of 0.003 with the Adam optimizer. In the light of the promising results obtained from this method, it is possible to say that the proposed method can be helpful in the early diagnosis in the upcoming waves of COVID-19
|