An adaptive adjacency matrix-based graph convolutional recurrent network for air quality prediction.

In recent years, air pollution has become increasingly serious and poses a great threat to human health. Timely and accurate air quality prediction is crucial for air pollution early warning and control. Although data-driven air quality prediction methods are promising, there are still challenges in studying spatial-temporal correlations of air pollutants to design effective predictors. To address this issue, a novel model called adaptive adjacency matrix-based graph convolutional recurrent network (AAMGCRN) is proposed in this study. The model inputs Point of Interest (POI) data and meteorological data into a fully connected neural network to learn the weights of the adjacency matrix thereby constructing the self-ringing adjacency matrix and passes the pollutant data with this matrix as input to the Graph Convolutional Network (GCN) unit. Then, the GCN unit is embedded into LSTM units to learn spatio-temporal dependencies. Furthermore, temporal features are extracted using Long Short-Term Memory network (LSTM). Finally, the outputs of these two components are merged and air quality predictions are generated through a hidden layer. To evaluate the performance of the model, we conducted multi-step predictions for the hourly concentration of PM2.5, PM10 and O3 at Fangshan, Tiantan and Dongsi monitoring stations in Beijing. The experimental results show that our method achieves better predicted effects compared with other baseline models based on deep learning. In general, we designed a novel air quality prediction method and effectively addressed the shortcomings of existing studies in learning the spatio-temporal correlations of air pollutants. This method can provide more accurate air quality predictions and is expected to provide support for public health protection and government environmental decision-making.

Fetal and neonatal outcome in severe alloimmunization managed with intrauterine transfusion: 18-year experience in a tertiary referral hospital in China.

Background: Hemolytic disease of the fetus and newborn (HDFN) due to red cell alloimmunization, is an important cause of fetal and neonatal morbidity and mortality. However, fetal and neonatal outcome of HDFN managed with intrauterine transfusion (IUT) in China are unknown. In addition, fetal and neonatal outcomes according to the type of maternal red cell alloantibodies involved and outcomes of hydrops fetalis are also unclear. Objectives: The objective of this study was to evaluate fetal and neonatal outcomes of severe red-cell alloimmunization treated by IUT, to compare the outcomes according to the type of antibody, and to investigate the perinatal and postnatal outcomes of hydrops fetalis due to red cell alloimmunization. Methods: A retrospective study of pregnancies affected by HDFN and managed with IUT at a tertiary care university hospital in China between January 2001 and December 2018 was performed. Fetal and neonatal outcomes were investigated, and comparison of outcomes depending on the type of antibody and comparison of outcome between hydrops fetalis and fetuses without hydrops were also conducted. Results: 244 IUTs were performed in 81 fetuses from 80 pregnancies. Anti-RhD was the major etiology of HDFN requiring IUT (71.6%). The fetal survival rate was 90.1%. The survival rate of the hydropic fetuses was significantly lower than those of the non hydropic fetuses (61.2% vs. 95.6%) (P = 0.002**). Compared with non hydropic fetuses, hydropic fetuses had significantly lower gestational age and lower hemoglobin level at first IUT. The neonatal survival rate was 98.6%. Exchange transfusions were required in 26% of the neonates. 30.1% of neonates had late anemia and required top-up transfusions, and hydropic fetuses required more late top-up transfusions than fetuses without hydrops. No significant difference in fetal and neonatal outcomes was found among the four subgroups stratified by the antibody involved. Conclusion: Our study demonstrates that IUT is an effective and safe therapy for severe HDFN at our institution. Early detection and treatment of hydrops is critical for perinatal outcomes. Particular attention should be paid to late postnatal anemia in affected neonates and top-up transfusion is still commonly needed.

CaMK4 Promotes Acute Lung Injury Through NLRP3 Inflammasome Activation in Type II Alveolar Epithelial Cell.

Background: Type II alveolar epithelial cell (AEC II), in addition to its roles in maintaining lung homeostasis, takes an active role in inflammatory response during acute lung injury (ALI). Ca2+/calmodulin-dependent protein kinase IV (CaMK4) activated by Ca2+/calmodulin signaling, has been implicated in immune responses. This study was to investigate the roles of CaMK4 in the development of ALI and the underlying mechanisms. Methods: CaMK4 inhibitor KN-93 was used to investigate the effects of CaMK4 on NLRP3 inflammasome activation. The effects of KN-93 on disease development of lipopolysaccharide (LPS)-induced ALI were also evaluated. The role of CaMK4 on NLRP3 inflammasome activation was explored in human AEC II cell line A549 using KN-93 or CaMK4 siRNA. NLRP3 inflammasome activation was measured by histology immunofluorescence and Western blot. IL-1ß and IL-18 were measured by ELISA. Results: Phosphorylation of CaMK4 and the expression of NLRP3 and Caspase-1 p20 were increased in the lungs of LPS-induced ALI mice, which was suppressed by KN-93 as measured by Western blot. Further, the activation of NLRP3 inflammasome was detected in AEC II from patients with acute respiratory distress syndrome (ARDS) and LPS-induced ALI mice. In vitro, inhibition or silencing CaMK4 in AEC II significantly inhibited NLRP3 inflammasome activation, resulting in reduced IL-1ß production. The inhibition of NLRP3 inflammasome and decreased IL-1ß/IL-18 production by KN-93 led to reduced inflammatory infiltration and ameliorated lung injury in LPS-induced ALI mice. Conclusion: CaMK4 controls the activation of NLRP3 inflammasome in AEC II during LPS-induced ALI. CaMK4 inhibition could be a novel therapeutic approach for the treatment of ALI.

A Novel Air Quality Early-Warning System Based on Artificial Intelligence.

The problem of air pollution is a persistent issue for mankind and becoming increasingly serious in recent years, which has drawn worldwide attention. Establishing a scientific and effective air quality early-warning system is really significant and important. Regretfully, previous research didn't thoroughly explore not only air pollutant prediction but also air quality evaluation, and relevant research work is still scarce, especially in China. Therefore, a novel air quality early-warning system composed of prediction and evaluation was developed in this study. Firstly, the advanced data preprocessing technology Improved Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (ICEEMDAN) combined with the powerful swarm intelligence algorithm Whale Optimization Algorithm (WOA) and the efficient artificial neural network Extreme Learning Machine (ELM) formed the prediction model. Then the predictive results were further analyzed by the method of fuzzy comprehensive evaluation, which offered intuitive air quality information and corresponding measures. The proposed system was tested in the Jing-Jin-Ji region of China, a representative research area in the world, and the daily concentration data of six main air pollutants in Beijing, Tianjin, and Shijiazhuang for two years were used to validate the accuracy and efficiency. The results show that the prediction model is superior to other benchmark models in pollutant concentration prediction and the evaluation model is satisfactory in air quality level reporting compared with the actual status. Therefore, the proposed system is believed to play an important role in air pollution control and smart city construction all over the world in the future.