Search

Coronavirus disease-19 (COVID-19) is caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and is currently a major cause of intensive care unit (ICU) admissions globally. The role of machine learning in the ICU is evolving but currently limited to diagnostic and prognostic values. A decision tree (DT) algorithm is a simple and intuitive machine learning method that provides sequential nonlinear analysis of variables. It is simple and might be a valuable tool for bedside physicians during COVID-19 to predict ICU outcomes and help in critical decision-making like end-of-life decisions and bed allocation in the event of limited ICU bed capacities. Herein, we utilized a machine learning DT algorithm to describe the association of a predefined set of variables and 28-day ICU outcome in adult COVID-19 patients admitted to the ICU. We highlight the value of utilizing a machine learning DT algorithm in the ICU at the time of a COVID-19 pandemic. This was a prospective and multicenter cohort study involving 14 hospitals in Saudi Arabia. We included critically ill COVID-19 patients admitted to the ICU between March 1, 2020, and October 31, 2020. The predictors of 28-day ICU mortality were identified using two predictive models: conventional logistic regression and DT analyses. There were 1468 critically ill COVID-19 patients included in the study. The 28-day ICU mortality was 540 (36.8 %), and the 90-day mortality was 600 (40.9 %). The DT algorithm identified five variables that were integrated into the algorithm to predict 28-day ICU outcomes: need for intubation, need for vasopressors, age, gender, and PaO2/FiO2 ratio. DT is a simple tool that might be utilized in the ICU to identify critically ill COVID-19 patients who are at high risk of 28-day ICU mortality. However, further studies and external validation are still required. [ABSTRACT FROM AUTHOR]

Abstract In the present study a fibrinolytic enzyme producer was isolated and identified as Bacillus subtilis using 16S rDNA sequencing. Central Composite Design was used for optimization of enzyme production using fodder yeast as a cost effective growth medium. The obtained results revealed that fodder yeast concentration, incubation temperature, aeration level followed by yeast extract concentration and incubation period are significant factors affect the enzyme production yield by the tested organism. Optimum levels of the selected variables were 3.05% fodder yeast, 0.71% yeast extract, initial pH 7,20% aeration level, 3.2% inoculum size (16 × 106CFU), 36.7 °C incubation temperature and 4 days incubation period. At these conditions the predicted enzyme activity was 18.9 U/ml and the practical enzyme activity was 16.6 U/ml which revealed that the model was valid by 87.83%. The results were discussed in the light of possible application as a thrombolytic agent. [ABSTRACT FROM AUTHOR]

• A full-bore rupture in the form of a fast-running ductile fracture has occurred at an 18-inch offshore pipeline in the subsea portion. • The visual inspection and ROV survey both revealed a conspicuous dent on the outer surface of the pipeline. • The visual examination revealed that the crack originated at a sharp weld toe angle (149°) at the outer surface. • The rupture origin was a fatigue pre-crack of 0.078-inch depth and 6.7-inch length that propagated at a sharp weld toe angle during the pipeline service life. • The stress analysis revealed that the burst was not due to the operation pressure, but rather due to stresses introduced by a sharp tool dragging over the bare pipeline surface. In the current work, a full-bore rupture in the form of a fast-running ductile fracture has occurred at an 18-inch offshore pipeline in the subsea portion. The pipeline was constructed from longitudinal seam welded pipe conforming to API 5L Grade X42. A subsea survey using the remote operation vehicle showed a pipeline rupture at a depth of 2756 in. from the sea surface. Shiny scratch marks on the outer surface of the pipeline at the rupture location were observed as well. The visual examination revealed that the crack originated at a sharp weld toe angle (149°) at the outer surface. The crack originated in the HAZ region and propagated along the longitudinal seam weld. The rupture origin was a fatigue pre-crack of 0.078-inch depth and 6.7-inch length that propagated at a sharp weld toe angle during the pipeline service life. The stress analysis revealed that the burst was not due to the operation pressure, but rather due to stresses introduced by a sharp tool dragging over the bare pipeline surface. The tool introduced huge external stresses to the pipeline. Both of the internal pressure and the external stresses contributed to the crack propagation during the rupture. [ABSTRACT FROM AUTHOR]

• The high salinity broke down the passive layer of the DSS. • The high level of CO 2 accelerated the corrosion rate of the de-passivated locations. • The selective dissolution of the ferrite was accelerated in part due to the enriched-Ni filler metal. • The excessive heat input during the welding played a role in the degradation of the HAZ zones. • The low levels of nitrogen impairing the pitting and crevice corrosion resistance. • The relatively high service temperature accelerated substantially the damage. In the current study, a 2205 duplex stainless-steel gas flowline with a diameter of 6-inch experienced a leak at two different joints after two years of service. The retrieved failed joints showed severe corrosion damage in the weld and heat-affected zones. This investigation revealed that the damage occurred in the flow line at the 4 to 8 o'clock position (i.e., angle positions of 120° to 240°) in contact with the high salinity water, and negligibly occurred in the top part of the pipe exposed to the dry gasses only. The high salinity broke down the passive layer of the steel. Furthermore, the high carbon dioxide level accelerated the corrosion rate at the de-passivated locations. The selective dissolution of the ferrite was partially accelerated due to the enriched-Ni filler metal. Moreover, the excessive heat input resulted in a coarse-grained heat-affected zone. The low nitrogen levels in the pipes and the flanges impaired the pitting and crevice corrosion resistance of the joints. The relatively high temperature also played a role in accelerating the selective dissolution of metal at the damaged locations. [ABSTRACT FROM AUTHOR]

• We propose a Behavior Regularized Prototypical Network (BR-ProtoNet) for few-shot image classification in semi-supervised scenarios. • BR-ProtoNet enables metric learning to benefit from readily-available unlabeled data. • We construct complementary constraints to regularize the model's behavior over the neighborhoods of training instances and along the interpolation paths among them. • The constructed regularization encourages the learnt embedding space to possess the property of proximity preservation. We propose a Behavior Regularized Prototypical Network (BR-ProtoNet) for few-shot image classification in semi-supervised scenarios. To learn a generalizable metric, we exploit readily-available unlabeled data and construct complementary constraints to regularize the model's behavior. Specifically, we match the label spaces between each episode and the whole training set. The predictions on the unlabeled data over different episodes can be aggregated to capture more reliable category information. We further construct new instances via adversarial perturbation and interpolation. These instances regularize the model's behavior over the neighborhoods of the original ones and along the interpolation paths among them. In addition, they ensure the learnt embedding space possesses the property of proximity preservation. The regularization of these aspects is incorporated into the optimization process of BR-ProtoNet on partially labeled data. We have conducted thorough experiments on multiple challenging benchmarks. The results suggest that the metric learning can significantly benefit from the proposed regularization, and thus leading to the state-of-the-art performance in semi-supervised few-shot image classification. [ABSTRACT FROM AUTHOR]