Detecting Covid-19 And Other Pneumonia Diseases Using Shufflent Cnn

Due to the exponential growth of the novel coronavirus disease (COVID-19), scientists are still exploring for precise and rapid diagnostic methods to detect potential COVID-19-caused lung infections. Thus, for the purpose of automatic COVID-19 detection, we proposed a practical and effective deep learning model based on chest X-ray image classification that accurately discriminates COVID-19 cases from other pneumonia and healthy chest cases. A novel convolutional neural network algorithm, ShuffleNet, was used to extract discriminative features from a variety of chest X-ray (CXR) images, including those with bacterial pneumonia, viral pneumonia, COVID-19, and normal CXR scans, before classifying them using the support vector machine (SVM) technique. The model has been trained on three various datasets. The performance of the model is measured using different criteria: accuracy, sensitivity, and precision. The results show that the proposed model outperformed SOA studies. In the first dataset, the result of thee-class classification has demonstrated higher accuracy, sensitivity, and precision (98%, 100%, and 100%, respectively) in classifying COVID-19 than state-of-art models. In the other two datasets, the model has achieved four-class classifications and performed accuracy of 95.58% and 91.8, respectively. It also achieved high sensitivity (95.5 and 95.6 in consequently classifying COVID-19).