Computer-Aided Diagnosis of COVID-19 Radiography using deep transfer deep learning model TLDeNet

Computer-Aided Diagnosis (CAD) is applied in the medical analysis of X-ray images widely. Due to the COVID-19 pandemic, the speed of COVID-19 detection is slow, and the workforce is scarce. Therefore, we have an idea to use CAD to diagnose COVID-19 and effectively respond to the pandemic. Recent studies show that convolutional neural network (CNN) is an appropriate technique for medical image classification. However, CNN is more suitable for datasets with many images, such as ImageNet. Medical image classification relies on doctors to label medical images, so obtaining large-scale medical image data sets is a time-consuming, costly, and unrealistic task. The method of data augmentation for a limited medical dataset can be used to increase the number of images. However, this technology will produce many repeated images, which will easily lead to the overfitting problem of CNN. In the case of a limited number of radiological images, transfer learning is a practical and effective method which can help us overcome the overfitting problem of ordinary CNN by transferring the pre-Trained models on large datasets to our tasks. The proposed model is DenseNet based deep transfer learning model (TLDeNet) to identify the patients into three classes: COVID-19, Normal or Pneumonia. We then analyzed and assessed the performance of our model on COVID-19 X-ray testing images by performing extensive experiments. It is finally demonstrated that the proposed model is superior to other deep transfer learning models according to comparative analyses. The Grad-Cam method is finally applied to interpret the convolutional neural network, revealing that our proposed model focuses on the similar region of the X-ray images as doctors. © 2022 IEEE.

High-resolution satellite-based PM2.5 concentration data acquired during the COVID-19 outbreak throughout China: Model, variations and reasons

High spatial resolution and broad spatial coverage data on fine particulate matter (PM2.5) are of great significance to estimating the exposure to PM2.5. However, the data is currently very limited worldwide. In addition, the COVID-19 pandemic in China, starting in January 2020, have led to significant variations in the PM2.5 concentrations. To identify the variations and causes of PM2.5 concentrations before and after the COVID-19 pandemic from 23 January to 24 March during 20182020, a geographically weighted regression model with a 1 km spatial resolution covering all of mainland China was developed. The overall R and RMSE values of the model cross validation were 0.91 and 17.19 g/m3, respectively, indicating that the model performed satisfactorily in estimating the PM2.5 values. Then, based on the satellite-based PM2.5 values, the results show that the PM2.5 values fluctuated significantly across mainland China before and after the COVID-19 outbreak. Additionally, the mean PM2.5 values decreased by 5.41 g/m3 in 2020 compared to 2019. In Hubei Province, the mean PM2.5 values increased by 1.85 g/m3 in 2019 compared to 2018, whereas they dramatically decreased by 23.18 g/m3 in 2020 compared to 2019. Finally, the results show that anthropogenic factors were primarily responsible for the variations in the PM2.5 concentrations in Heilongjiang, Jilin, and Liaoning provinces;whereas, both meteorological and anthropogenic factors were responsible for the variations in Hubei, Henan, Anhui, Shandong, and Jiangsu provinces during the study period. These results provide an important reference for the future development of air pollution control policies in China. Author

Summary of perioperative hemodynamic management of 19 lung transplant recipients

Objective: To summarize and discuss the experience of perioperative hemodynamic management of lung transplant recipients. Methods: A total of 19 lung transplant recipients from December 2016 to December 2020 were investigated in Renmin Hospital of Wuhan University, all of which were transferred to the intensive care unit for further monitoring and treatment, and their clinical data were retrospectively analyzed. Results: The 19 lung transplant recipients included 5 cases of chronic obstructive pulmonary disease (COPD), 5 cases of idiopathic pulmonary fibrosis (IPF), 4 cases of pneumoconiosis, 2 cases of bronchiectasis, 1 case of later lung fibrosis associated with COVID-19, 1 case of connective tissue disease-related pulmonary fibrosis, and 1 case of Kartagener syndrome. Twelve cases adopted double lung transplantation, while seven cases reveived unilateral lung transplantation (4 cases of left single lung transplantation and 3 cases of right single lung transplantation). There were 6 deaths during the perioperative period. One case died of multi-drug resistant bacteria infection, one case died of circulatory failure caused by active thoracic hemorrhage post-operation, the third case died of intraoperative cardiac arrest, and the other 3 cases were given up because of multiple organs failure. The remaining 13 cases were cured and discharged. Of the 19 recipients, 14 received vasopressors. The total and daily fluid output of the recipients in 3 postoperation days were greater than the input volume (P<0.05). Conclusion: Lung transplantation is an effective method for the treatment of end-stage lung disease. The hemodynamic management is a keypoint during perioperative period. It is import to maintain the blood volume as low as possible under the premise of systemic perfusion, limit the amount of fluid, choose albumin or plasma to increase the colloidal osmotic pressure, and strengthen the maintenance of right heart function. These abobe measures may improve the prognosis of lung transplant recipients. © 2021, Editorial Board of Medical Journal of Wuhan University. All right reserved.

Study on soybean potential productivity and food security in China under the influence of COVID-19 outbreak

The COVID-19 outbreak that became a global pandemic in early 2020 is starting to affect agricultural supply chains and leading to a rapid rise in global food prices. As many grain exporting countries announced a ban on grain exports, food security issues in China have attracted a significant international attention. Based on the Suitability Distribution Model and Soybean-Cereal Constraint Model, we explored the relationship between soybean production potential and food security. We calculated that the soybean potential planting area in China is 164.3 million ha. If the outbreak prevents China from importing soybeans, soybean planting area will need to be increased by 6.9 times to satisfy the demands. In the meantime, cereal self-sufficiency rate will drop to 63.4%, which will greatly affect food security. Each additional unit of soybean production will reduce 3.9 units of cereal production, and 1% increase in the self-sufficiency rate of soybean will result in a 0.63% drop in the self-sufficiency rate of cereal. Without sacrificing the self-sufficiency rate of cereal, the self-sufficiency rate of soybean is limited to 42%. Consequently, China will still need to import more than 68% of the current import volume of soybean. Although in the short term, the outbreak will not affect food security in China, as soybean imports decrease, insufficient supply of soybeans will affect people's quality of life. To prevent the impact of the COVID-19 outbreak, China should increase soybean stocks and strengthen international cooperation. In the long term, increasing the self-sufficiency rate is a fundamental solution to solving soybean import dependency. The key to increasing soybean cultivation is by making soybean cultivation profitable and by building a sustainable soybean planting chain.

Predictors of progression from moderate to severe coronavirus disease 2019: a retrospective cohort.

OBJECTIVE: Most cases of coronavirus disease 2019 (COVID-19) are identified as moderate, which is defined as having a fever or dry cough and lung imaging with ground-glass opacities. The risk factors and predictors of prognosis in such cohorts remain uncertain. METHODS: All adults with COVID-19 of moderate severity diagnosed using quantitative RT-PCR and hospitalized at the Central Hospital of Wuhan, China, from 1 January to 20 March 2020 were enrolled in this retrospective study. The main outcomes were progression from moderate to severe or critical condition or death. RESULTS: Among the 456 enrolled patients with moderate COVID-19, 251/456 (55.0%) had poor prognosis. Multivariate logistic regression analysis identified higher neutrophil count: lymphocyte count ratio (NLR) on admission (OR 1.032, 95% CI 1.042-1.230, p 0.004) and higher C-reactive protein (CRP) on admission (OR 3.017, 95% CI 1.941-4.690, p < 0.001) were associated with increased OR of poor prognosis. The area under the receiver operating characteristic curve (AUC) for NLR and CRP in predicting progression to critical condition was 0.77 (95% CI 0.694-0.846, p < 0.001) and 0.84 (95% CI 0.780-0.905, p < 0.001), with a cut-off value of 2.79 and 25.95 mg/L, respectively. The AUC of NLR and CRP in predicting death was 0.81 (95% CI 0.732-0.878, p < 0.001) and 0.89 (95% CI 0.825-0.946, p < 0.001), with a cut-off value of 3.19 and 33.4 mg/L, respectively. CONCLUSIONS: Higher levels of NLR and CRP at admission were associated with poor prognosis of individuals with moderate COVID-19. NLR and CRP were good predictors of progression to critical condition and death.