No abstract

BackgroundThrough the coronavirus disease 2019 (COVID-19) pandemic, portable radiography was particularly useful for assessing and monitoring the COVID-19 disease in Vietnamese field hospitals. It provides a convenient and precise picture of the progression of the disease. The purpose of this study was to evaluate the predictive value of chest radiograph reporting systems (Brixia and total severity score (TSS)) and the National Early Warning Score (NEWS) clinical score in a group of hospitalized patients with COVID-19. MethodsThis retrospective cohort study used routinely collected clinical data from polymerase chain reaction (PCR)positive severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) patients admitted to Field Hospital District 8, Ho Chi Minh City, Vietnam, from August 2021 to September 2021. The initial chest radiographs were scored based on the TSS and Brixia scoring systems to quantify the extent of lung involvement. After the chest radiograph score was reported, two residents calculated the rate of all-cause in-hospital mortality with the consultation of expert radiologists. In this study, NEWS2 scores on hospital admission were calculated. The gradient boosting machines (GBMs) and Shapley additive exPlanations (SHAP) were applied to access the important variable and improve the accuracy of mortality prediction. The adjusted odds ratio for predictor was presented by univariate analysis and multivariate analysis. ResultsThe chest X-rays (CXRs) at the admission of 273 patients (mean age 59 years +/-16, 42.1% were male) were scored. In the univariate analysis, age, vaccination status, previous disease, NEWS2, a saturation of peripheral oxygen (Sp02), the Brixia and TSS scores were significant predictors of mortality (p-value < 0.05). In multivariate analysis, there were statistically significant differences in mortality between age, Sp02, Brixia score, and patients with previous diseases were independent predictors of mortality and hospitalization.A gradient boosting machine was performed in the train data set, which showed that the best hyperparameters for predicting the mortality of patients are the Brixia score (exclude TSS score). In the top five predictors, an increase in Brixia, age, and BMI increased the logarithmic number of probability clarifying as death status. Although the TSS and Brixia scores evaluated chest imaging, the TSS score was not essential as the Brixia score (rank 6/11). It was clear that the BMI and NEWS2 score was positively correlated with the Brixia score, and age did not affect this correlation. Meanwhile, we did not find any trend between the TSS score versus BMI and NEWS2. ConclusionWhen integrated with the BMI and NEWS2 clinical classification systems, the severity score of COVID-19 chest radiographs, particularly the Brixia score, was an excellent predictor of all-cause in-hospital mortality.

No abstract

Produced water is the largest by-product of oil and gas production. At off-shore installations, the produced water is typically reinjected or discharged into the sea. The water contains a complex mixture of dispersed and dissolved oil, solids and inorganic ions. A better understanding of its composition is fundamental to (1) improve environmental impact assessment tools and (2) develop more efficient water treatment technologies. The objective of the study was to screen produced water sampled from a producing field in the Danish region of the North Sea to identify any containing organic compounds. The samples were taken at a test separator and represent an unfiltered picture of the composition before cleaning procedures. The analytes were isolated by liquid-liquid extraction and derivatized using a silylation reagent to increase the volatility of oxygenated compounds. The final extracts were analyzed by comprehensive multi-dimensional gas chromatography coupled to a high-resolution mass spectrometer. A non-target processing workflow was implemented to extract features and quantify the confidence of library matches by correlation to retention indices and the presence of molecular ions. Approximately 120 unique compounds were identified across nine samples. Of those, 15 were present in all samples. The main types of compounds are aliphatic and aromatic carboxylic acids with a small fraction of hydrocarbons. The findings have implications for developing improved environmental impact assessment tools and water remediation technologies.

No abstract