Revolutionizing Covid-19 Diagnosis by Deep Learning Using X-Ray Images

The outbreak of the COVID-19 pandemic has presented significant challenges to global healthcare systems, necessitating a timely and accurate detection of COVID-19 cases. In this research paper, the authors have proposed an adapted Detrac model (a novel deep learning-based approach) designed to classify input images into three distinct categories: pneumonia, COVID, normal) called DeepEnTraCT for COVID-19 detection using chest X-ray images. This innovation aims to improve the precision and efficiency of COVID-19 detection models by incorporating advanced techniques in feature extraction, selection, and classification. The suggested model employs dual feature extraction techniques, leveraging a pre-trained AlexNet model to extract discriminative features from raw images and extracting statistical, GLCM, and PCA features. The authors perform feature selection using the Improved Salp Swarm Algorithm (ISSA), enabling the selection of essential elements and reducing dimensionality. In addition, the images are classified into normal, COVID-19, and pneumonia using the DeepEnTraCT model. Furthermore, they evaluate their results with that of conventional models and demonstrate better F-score performance, sensitivity, specificity, accuracy, and precision. A variety of MATLAB tests were carried out to evaluate the DeepEnTraCT model's performance utilizing a data set containing chest X-ray pictures from COVID-19, Normal, and other virally induced pneumonia cases.