ABSTRACT
Corona Disease Virus (COVID-19) is a rapidly spreading contagious viral disease that causes respiratory contaminations and is currently generating a worldwide medical crisis. It has caused a massive influence on people's lives, general well-being, and the global economy. Henceforth, it is critical to straightaway analyze the positive cases in order to keep the illness from spreading further and to regard infected patients as fast as could really be expected. Both patients and specialists will be benefitted by the early recognizable capability of outrageous COVID-19 by utilizing chest CT to examine biomedical images. RT-PCR (switch record polymerase chain response) based tests help to identify COVID-19, which has numerous limits. In this work, different CNN based Classifier model methodologies are utilized to follow the presence of COVID-19 from chest CT filter images of patients. In true indicative situations, a profound CNN-based methodology could be amazingly valuable in accomplishing quick COVID-19 testing. By utilizing irregularity data obtained from sifted images, image expansion enhances the number of profitable models for creating the CNN model. The proposed model has a grouping exactness of 95% for CT examines utilizing this strategy. With picture expansion, CT check pictures have an affectability of 94.78%and a particularity of 95.98%. The trial results were contrasted with ResNet-18, ResNet-50, and VGG-16 models, with freely available datasets containing CT images. © 2021 IEEE.