| dc.description.abstract | SARS-CoV-2's COVID-19 pandemic has quickly spread over the world, inflicting a sizable number of illnesses and fatalities. Stopping the virus's spread depends on correctly and rapidly identifying infected people. Although RT-PCR assays, for example, are thought to be the most accurate way to identify COVID- 19, their cost and availability may be restricted in places with limited resources. In this study, we propose some deep-learning methods for predicting COVID-19 detection using chest X-ray images. Chest X-ray imaging has become an essential diagnostic tool in the management of COVID-19, as it is non-invasive, widely available, and cost-effective. However, the interpretation of chest X-rays for COVID-19 detection can be challenging, as the radiographic features of COVID-19 pneumonia can be subtle and overlap with other respiratory diseases. In this study, the performance of different deep learning models, notably VGG16, VGG19, DenseNet121, and Resnet50, was examined for their ability to distinguish between coronavirus pneumonia and cases of pneumonia. 4649 chest X-ray images of patients with pneumonia (3526) and COVID-19 (1123) were employed in the study, and performance measures were used to assess each model. Confusion metrics were also used to evaluate each model's performance. The study's findings demonstrated that DenseNet121 performed better than the competing models, with an accuracy rate of 99.78%. | en_US |
