Enhancing COVID-19 Diagnosis with Automated Reporting using Preprocessed Chest X-Ray Image Analysis based on CNN

Abstract

The ongoing COVID-19 pandemic has caused a global health crisis, and accurate diagnosis and early detection are essential for successful management of the outbreak. Convolutional neural networks(CNNs) and preprocessed chest X-ray pictures are the two main components of the unique proposed method for the identification of COVID-19, which is presented in this study. Image enhancement and segmentation are performed during the pre-processing stage. These operations increase the overall quality and contrast of the pictures, which in turn makes it simpler for the CNN to recognise significant aspects of the images. The CNN model was trained using a large dataset of pre-processed X-ray pictures that included both COVID-19 positive and negative instances. The dataset was used to train the model. In comparison to more conventional diagnostic approaches, and this strategy was successful in achieving high levels of accuracy, sensitivity, and specificity in the detection of COVID-19. Moreover, this model designed an automated reporting system that saves time and costs by providing healthcare providers with diagnostic reports that are both prompt and accurate. This research demonstrates the viability of using CNNs and pre-processed X-ray images for the purpose of early identification of COVID-19 and offers an important resource for the efficient management of this worldwide health concern.