Published Paper Details:

The SARS-CoV2 virus caused a new corona virus that started in Wuhan, China and spread around the world. Millions of people were infected as a result of the virus's widespread spread. Early discovery of the virus is critical for the patient's complete recovery, but late detection can be fatal. Because the virus's symptoms are similar to those of the flu, it's tough to spot. This research seeks to develop an automated approach for detecting Covid19 from virus-infected chest X-ray pictures. The suggested method makes use of a dataset that includes non-infected participants as well as patients with pneumonia and Covid19 virus infection. For feature extraction, local binary patterns with variations in their input parameters are used. Several machine learning methods and ensembles of these individual models are used to classify the generated feature sets. Experiment results are acquired using 10-fold cross validation testing. To compare performance, the evaluation metrics accuracy, positive predictive value (PPV), sensitivity, and f-measure are utilised. The results reveal that the RTree-RForest-KNN ensemble delivers the best classification performance, whereas ensemble models outperform most individual classifiers. When comparing the input parameters of the LBP, the best performance is given by parameters R=6 (P=48) and R=7 (P=56) in the suggested Covid19 identification approach from chest X-Ray pictures for the average of metrics for 10-fold cross validation.