Study and analysis of various COVID-19 prediction techniques using CT images: A challenging overview

Abstract

coronavirus disease (COVID-19) has scattering quickly across a globe due to its exceedingly infectious natural world and is affirmed as epidemic by World Health organization (WHO). This deadly disease has depicts the globe to a standstill. From a breakdown economy to increasing bereavement charges, this viral disease is infectingeverybody. It is significant to hold the increase to diminish the danger of affection. This can be carried outby means of extensive testing and tracing of contacts. In the detection of COVID-19, three major screening processes are utilized, which include chest X-Ray (CXR), Computed Tomography (CT) as well as Reverse TranscriptasePolymerase Chain Reaction (RT-PCR). To struggle against a rapid increase of coronavirus, analysis of CT clinical images engage a vital part in precise diagnostic. Hence, this survey paper overviews some techniques consequent to a detection of COVID-19 utilizing CT images. This survey assess 25 research papers concentrated on COVID-19 prediction utilizing CT images and presented method-wise overviews, like deep learning-based techniques, optimization based methods, transfer learning based techniques, and machine learning based approaches. An evaluation takes part in a review based on cataloging probe techniques, tools etemployed datasets, publication year, and evaluation metrics for the prediction of COVID-19. At last, the troubles and demerits of reviewed methods are explicated, which motive probers for introducing new effective methods for predicting COVID-19 by wielding images of CT.