Deep learning for Covid-19 classification using CT scan slices of lung

ABSTRACT

Since the rapid spreading of covid-19 in 2019 in the whole world, it was conceded in 2020 as a pandemic. The long timeline of PCR tests and lack of test tool kits in many hospitals leads to fast infection according to the slow diagnosis. Various experiences of radiologists cause deferent in accurately detection lessons. This research suggested and designed a model based on utilizing the deep learning (DL) algorithms to detect the infection of covid-19 patients. Transfer learning VGG16 has been manipulated and used to solve the problem. Manipulating on VGG16 has been accomplished to achieve acceptable accuracy. The tuning on the last three layers of VGG16 architecture (dense layers) by replacing them with two layers (flatten layer and dense layer). The dense layer that is added deals with binary classification problems depending on the sigmoid function. This tuning serves the current study by speeding up the prediction of the model and also increasing the accuracy. A large COVID-19 CT scan slice dataset has been used to train and test the model. The result of testing reached 99.7% with a loss of 0.0085 and a validation loss of 0.0162. The obtained result proved that the system can help the radiologist accommodate the pandemic. © 2022 IEEE.