The main structure of the paper is as follows: Section II discusses the advanced work on using X-ray images for COVID-19. Therefore, in this work, we propose a method based on feature fusion that can more accurately distinguish among healthy, common pneumonia, and COVID-19 patients using chest X-ray images. Compared with other advanced methods, the results show that our model can classify chest X-ray data with high accuracy. We conducted two-category work for healthy and pneumonia patients and three-category work for healthy, common pneumonia, and COVID-19 patients. 1, it can assist doctors in achieving more accurate classification and diagnosis. These differences or subtle features can be detected by artificial intelligence. Previous studies have shown that CXR images have specific differences in the imaging manifestations of common pneumonia and COVID-19. Therefore, CXR, as a sensitive method to detect COVID-19 as well as other chest problems, plays an integral role in the early diagnosis and treatment of the disease. Although the antigen test has a relatively fast detection speed, its sensitivity is poor. ![]() LAMP technology has high sensitivity, fast reaction rate and strong specificity, but the design of primers is very complicated, and it is easy to produce nonspecific amplification. Although the specificity of RT–PCR is sufficiently high for COVID-19, its sensitivity is relatively low in detecting COVID-19. Real-time polymerase chain reaction (RT–PCR), loop-mediated isothermal amplification (LAMP), antigen testing and other methods can be used to detect COVID-19. The common clinical symptoms are mainly respiratory symptoms, and some patients may have gastrointestinal symptoms. The characteristics of COVID-19 are diverse and unpredictable. It was named COVID-19 by the World Health Organization (WHO) in February 2020 around March 2020, the World Health Organization announced that the disease has affected the whole world and is a global pandemic disease. The outbreak was declared a Public Health Emergency of International Concern on 30 January 2020. The pandemic of global concern caused by COVID-19 has also brought enormous challenges to governments and the healthcare industry. X-ray is one of the most common radiological examination methods for screening and diagnosing chest diseases, as well as the main means of classifying and screening pneumonia, tuberculosis and breast cancer, and is a painless and noninvasive examination method suitable for high populations with relatively low costs. Radiologists can use CXR features to determine the type of pneumonia and the underlying pathogenesis. Chest X-rays (CXRs) play an important role in patient care. Pneumonia-type illnesses are more contagious during the flu season. Therefore, the use of deep learning and feature fusion technology in the classification of chest X-ray images can become an auxiliary tool for clinicians and radiologists. The experimental results show that the proposed model has good results in this work. The average accuracy for three category classification can reach 97.3%. The average accuracy of our model in detecting binary classification can reach 98.0%. A residual network (ResNet) is used to segment effective image information to quickly achieve accurate classification. This paper adds an attention mechanism (global attention machine block and category attention block) to the model to extract deep features. In this study, we propose a chest X-ray image classification method based on feature fusion of a dense convolutional network (DenseNet) and a visual geometry group network (VGG16). ![]() For this high-speed infectious disease, the application of X-ray to chest diagnosis plays a key role. Since December 2019, the novel coronavirus disease (COVID-19) caused by the syndrome coronavirus 2 (SARS-CoV-2) strain has spread widely around the world and has become a serious global public health problem.
0 Comments
Leave a Reply. |