Neural network-based strategies for automatically diagnosing of COVID-19 from X-ray images utilizing different feature extraction algorithms

The COVID-19 pandemic has had an obliterating impact on the health and well-being of the worldwide populace. It has recently become one of the most severe and acute diseases and has spread globally. Therefore, an automated detection system should be implemented as the fastest diagnostic option to control the spread of COVID-19. This study aims to introduce neural network-based strategies for classifying and detecting COVID-19 early through image processing utilizing X-ray images. Despite the extensive use of directly fed X-ray images into the classifier, there is a lack of comparative analysis between the feature-based system and the direct imaging system in the classification of COVID-19 to evaluate the efficiency of feature extraction methods. Therefore, the proposed system represents introductory experiments of feature-based system using image Feature Extraction Algorithms [Texture, Grey-Level Co-occurrence Matrix (GLCM), Grey-Level Dependence Matrix (GLDM), Fast Fourier Transform (FFT), and Discrete Wavelet Transform (DWT)], Dimensionality Reduction Algorithms (Principal Component Analysis (PCA), Kernel Principal Component Analysis (KPCA), Sparse Autoencoder, and Stacked Autoencoder), Feature Selection Algorithms (Anova F-measure, Chi-square Test, and Random Forest), and Neural Networks [Feed-Forward Neural Network (FFNN) and Convolutional Neural Network (CNN)] on generated datasets consisting of Normal, COVID-19, and Pneumonia-infected chest X-ray images. Tenfold Cross-Validations are implemented during the classification process. A baseline model is created where no Feature Extraction Algorithm is implemented. Accuracy, sensitivity, specificity, precision, and F-measure metrics are utilized to assess classification performance. Finally, a comprehensive comparison of the success of the Feature-based system (Feature Extraction Algorithms, Dimensionality Reduction Algorithms, and Feature Selection Algorithms) in COVID-19 classification from X-ray images is made with the baseline model. The highest classification accuracy (95.44 ± 0.03%), sensitivity (96%), specificity (98%), precision (96%), and F-measure (96%) are achieved for Feature-based systems using Principal Component Analysis. The aim is to lay the foundation for the potential creation of a system that can automatically distinguish COVID-19 infection based on chest X-ray images using the Feature-based model.