Analysing COVID-19 Pneumonia Severity Through Chest X-Ray Radiography: A Computational Approach

Convolutional Neural Networks (CNN) and Recurrent Neural Networks (RNN) are used to compare how well COVID-19 pneumonia severity is assessed. The work employs deep learning algorithms on chest X-ray images to identify pneumonia site and measure severity. It leverages data from the RSNA Pneumonia Detection Challenge on Kaggle. Even with the latest developments in medical image processing, problems including computational complexity, X-ray picture variability, and precise localization of pneumonia-affected areas still exist. Using bounding box approaches, the methodology entails image recognition and the calculation of the severity percentage of lung opacity in images impacted by pneumonia. According to the experimental data, CNN outperformed RNN, which had a mean accuracy of 80.7225%, with a mean accuracy of 86.3650%. These results demonstrate how well CNN performs when assessing the severity of pneumonia from chest X-ray pictures, offering insightful information for further study and practical uses.