Optimising accuracy rate of genomic image representation of human coronavirus sequences for COVID-19 detection

Due to the significant mortality rate associated with the coronavirus disease 2019 (COVID-19), it is impossible to ignore this newly discovered illness that has an impact on healthcare on a worldwide scale. At this time, physicians are making use of pictures produced by computed tomography (CT) in order to aid them in recognising COVID-19 in its earlier stages. In this study, a COVID-19 diagnostic system is built with the help of a convolutional neural network (CNN) and stacked autoencoder. Before using the three different CT imaging methods to tell the difference between normal and COVID-19 cases. During the training phase of the deep learning model that was used, a demanding and large-scale CT image dataset was utilised. This allowed for accurate reporting of the model's ultimate performance. This model was correct 88.30% of the time, sensitive 87.65% of the time, and specific 87.97% of the time.