Comparative Analysis of Different Machine Learning Models Applied on Different Datasets Available for Diagnosis of Covid-19

Abstract

Since breaking out in Wuhan, China in the last days of 2019, the novel COVID-19 pandemic has done a great deal of damage to the mankind; whether it is economic damage, psychological or social one. It was declared a pandemic in March 2020. PCR (Polymerase Chain Reaction) which is used to diagnose COVID-19 patients usually takes 24-72 hours ranging in different countries. A new idea of diagnosing COVID-19 with the help of radiography images has surfaced which has taken research world by storm. There are different machine learning models developed with the help of historic data i.e., datasets which classify COVID-19 patients within a few minutes. As there are different publicly available datasets on which dozens of models are developed, we would like to perform comparative analysis of these datasets. This would help us to identify different aspects of these datasets.

Keywords

• COVID-19, CNN, CT scans, CXR, PCR, Pneumonia, Radiography