A Comparision of CT and MRI Screening for Covid-19 Diagnosis in Different Deep Learning Models

Abstract

A general pandemic is one of the most bizarre and dangerous corruptions in reliable time. The speed of COVID cases is rising rapidly all over the planet. Notwithstanding the way that antibodies have been made, fundamental learning (DL) structures have been shown as a significant procedure for clinical finding and in various fields. Deep learning is structured as a structure subject to convolution neural relationship with learning. This paper needs to do a general assessment on clinical pictures like PC tomography checks (CT channel) and Magnetic Resonance Imaging (MRI) through different monster learning structures. This assessment looks at concerning structures made for the COVID-19 examination through epic learning shows on VGG-16, Inception V3, DenseNet-169, Resnet-50, and VGG-19 models and gives bits of information and illuminating records to set up these neural affiliations. A close assessment is done by using CNN Models to compare whether CT or MRI performance which gives the best results accumulated and a better focal learning model for accreditation. The purpose of this paper is to ease clinically-prepared experts' minds and help them understand the methodology for key learning upgrades and how they can be unavoidably used to fight COVID-19.