ABSTRACT
At the present time, COVID-19 considers the pandemic that has most dangerous, representing a great threat to both, Communities and individuals especially scientist and researchers. Determine the infection in the early time is the only way to avoid getting worst especially for those who have respiratory diseases. The researchers working daily without stopping to get the possible solutions to rescue patients or who have the Symptoms of the disease. CT-Scans and X-rays for the lungs consider one of the most popular methods used by the researchers, where analyzing this image to detect the infection by this pandemic. However, it demands long time to examine each test, beside a large number of radiology specialists will be required. Oxygen data may be an interesting signal data samples that can be used to detect the COVID-19 via number of oxygen blood tests, in this research, detection of COVID-19 through some tests of oxygen signals is an essential goal. Tests like oxygen saturation (spO2), Pulse Oximetry (plusOX), Arterial blood gas (ABG) and Carboxyhaemoglobin (COHb) are the most common and important test for this purpose. In this research, a method based on deep neural network as well as machine learning methods using (stochastic gradient decision, decision tree, random forest, linear regression and k-nearest neighbour) was proposed for detecting COVID-19 through analysing the X-rays of patients which will look for visual indicators found in the chest radiography imaging of COVID-19 patients. Also proposed architecture can be applied to the oxygen data when it will be available (lack of well-structured data). © 2023 Author(s).
ABSTRACT
Corona virus's correct and accurate diagnosis is the most important reason for contributing to the treatment of this disease. Radiography is one of the simplest methods to detect virus infection. In this research, a method has been proposed that can diagnose disease based on radiography (X-ray chest) and deep learning techniques. We conducted a comparative study by using three diagnosis models;the first one was developed by using traditional CNN, while the two others are our proposed models (second and third models). The proposed models can diagnose the COVID-19 infection, normal cases, lung opacity, and Viral Pneumonia according to the four categories in the covid19 radiography dataset. The transfer learning technology had used to increase the robustness and reliability of our model, also, data augmentation was used for reducing the overfitting and to increase the accuracy of the model by scaling rotation, zooming, and translation. The third model showed higher training accuracy of 93.18% compared to the two other models that are dependent on using traditional convolution neural networks with an accuracy of 70.28% of the first model, while the accuracy of the second model that uses data augmentation with traditional convolution neural is 90.1%, while the testing accuracy models was 68.27% for the first model, 87.55% for the second model, and 86.03% for the third model. © 2022,International journal of online and biomedical engineering. All Rights Reserved.