Deep Learning-Based Lung Infection Detection Using Radiology Modalities and Comparisons on Benchmark Datasets in COVID-19 Pandemic

ABSTRACT

The SARS-CoV-2 (severe acute respiratory syndrome coronavirus) pandemic, also known as COVID-19 (coronavirus 2019), impacted humanity worldwide and significantly impacted the healthcare community. COVID-19 infection and transmission have resulted in several international issues, including health hazards. Sore throat, trouble breathing, cough, fever, weariness, and other clinical signs have been described. In SARS-CoV-2 patients, the most common infections are in the lungs and the gastric intestine. Lung infections may be caused by viral or bacterial infections, physical trauma, or inhalation of harmful particles. This research presents deep learning-based approaches for COVID-19 infection detection based on radiological images, prevention and therapy based on benchmark publicly available datasets. Finally, the analysis and findings explore evidence-based methodologies and modalities, leading to a conclusion and possible future healthcare planning. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.