RESUMEN
With the massive damage in the world caused by Coronavirus Disease 2019 SARS-CoV-2 (COVID-19), many related research topics have been proposed in the past two years. The Chest Computed Tomography (CT) scan is the most valuable materials to diagnose the COVID-19 symptoms. However, most schemes for COVID-19 classification of Chest CT scan are based on single slice-level schemes, implying that the most critical CT slice should be selected from the original CT volume manually. In this paper, a statistical hypothesis test is adopted to the deep neural network to learn the implicit representation of CT slices. Specifically, we propose an Adaptive Distribution Learning with Statistical hypothesis Testing (ADLeaST) for COVID-19 CT scan classification can be used to judge the importance of each slice in CT scan and followed by adopting the non-parametric statistics method, Wilcoxon signed-rank test, to make predicted result explainable and stable. In this way, the impact of out-of-distribution (OOD) samples can be significantly reduced. Meanwhile, a self-attention mechanism without statistical analysis is also introduced into the backbone network to learn the importance of the slices explicitly. The extensive experiments show that both the proposed schemes are stable and superior. Our experiments also demonstrated that the proposed ADLeaST significantly outperforms the state-of-the-art methods.
RESUMEN
Beginning in December 2019, coronavirus disease 2019 (COVID-19), due to 2019-nCoV infection, emerged in Wuhan and spread rapidly throughout China and even worldwide. Employing combined therapy of modern medicine and traditional Chinese medicine has been proposed, in which Ma Xing Shi Gan Decoction (MXSGD) was recommended as a basic prescription and applied widely in the clinical treatment of COVID-19. We investigated the underlying mechanism of MXSGD in treating COVID-19 utilizing the approaches of integrating network pharmacology. A total of 97 active ingredients of MXSGD were screened out, and 169 targets were predicted. The protein-protein interaction network exhibited hub targets of MXSGD, such as Heat shock protein 90, RAC-alpha serine/threonine-protein kinase, Transcription factor AP-1, Mitogen-activated protein kinase 1, Cellular tumor antigen p53, Vascular endothelial growth factor A, and Tumour necrosis factor. Gene Ontology functional enrichment analysis demonstrated that the biological processes altered within the body after taking MXSGD were closely related to the regulation of such processes as the acute inflammatory response, chemokine production, vascular permeability, response to oxygen radicals, oxidative stress-induced apoptosis, T cell differentiation involved in the immune response, immunoglobulin secretion, and extracellular matrix disassembly. KEGG enrichment analysis indicated that the targets of MXSGD were significantly enriched in inflammation-related pathways, immunomodulation-related pathways, and viral infection-related pathways. The therapeutic mechanisms of MXSGD on COVID-19 may primarily involve the following effects: reducing inflammation, suppressing cytokine storm, protecting the pulmonary alveolar-capillary barrier, alleviating pulmonary edema, regulating the immune response, and decreasing fever.