Background and Aim: The World Health Organization has some recommendations for early diagnosis of COVID-19 for appropriate treatment. In the present study, automatic and accurate diagnosis of COVID-19 by chest CT scan images using the deep learning method was performed.
Methods: The steps performed in this algorithm for segmentation and identification of images of healthy lungs and lungs affected by COVID-19 were a selection of appropriate images, preprocessing of images including noise reduction, extraction of image properties, finally segmentation and image classification using a combination of the deep learning method and water wave optimization algorithm to detect COVID-19 in lung images. All modeling was done based on Matlab software.
Results: By applying the water wave optimization algorithm to the deep learning algorithm, its accuracy improved by some 7% in the diagnosis of COVID-19. Therefore, the proposed algorithm with an average accuracy of 98% has a high ability to be used in the clinic for accurate and rapid diagnosis of COVID-19, which can be of great help to the medical staff.
Conclusion: In the diagnosis phase of COVID-19, artificial intelligence can be used to detect patterns of medical images taken by CT scan.