COVID-19 Classification With Healthcare Images Based on ML-DL Methods

Abstract

COVID-19 is the contagious ailment caused by Sars-Cov-2. This causative 2019-nCoV is a communication to the lines of millions of people. This study employs ML and DL epitomes to determine sickness along with predicting if a person is afflicted with the virus as the previous reports can examine the data pre-processing, feature extraction, classification, evaluation of experimental results to find advanced fact-finding directions around COVID-19 classification employing machine-deep approaches. The comparison shows that chest x-rays and CT are the most frequently used data in the diagnosis of COVID-19 rather than RT-PCR, and that the most-used test techniques were found to be insensitive and less beneficial after changing the limited number of datasets. This study suggests image preprocessing, exploratory data analysis, feature extraction (LBP), and other ML as well as DL classification methods. It attempts to minimise some of the issues that have been addressed for early identification for future work and studies.