Search

An improved RDN image super-resolution reconstruction network incorporating multi-scale residual dense network (MSRDN) was proposed to address the problem of inadequate use of image features in common single-image super-resolution (SISR) reconstruction networks. Firstly, the shallow features were extracted from the input low-resolution image. Then, the feature extraction module was jointly constructed using convolutional layers and local residual learning structures, and the multiplexed structure of the module at different scales was used to fully extract the multi-scale detail features of the image. Next, a top-down and bottom-up feature fusion module was constructed to fully fuse and correlate the collected multi-scale features to construct image features with richer detail information. Finally, the extracted features were sent to the image reconstruction module for the final super-resolution image reconstruction. The experimental results show that the proposed MSRDN network exhibits better visual results compared to networks such as SMSR on the three super-resolution benchmark sets of Set5, Set14 and BSD100, with the peak signal to noise ratio (PSNR) the obtained peak singal improved by 0.8 dB on average, and structural similarity (SSIM) by an average of 0.02.

In general, studies on green innovation practices have primarily taken an innovation view to study the costs and benefits to firms. What has been less examined, however, is how green training spurs the intentions and advantages of green innovation. Thus, this paper explores how and when green training influences the likelihood of firms to achieve sustainability performance by using organizational learning theory. Using a dataset of 231 Chinese manufacturing firms, our work shows that green training is positively correlated to corporate sustainability performance via the indirect effect of green innovation behavior. We also find an increasingly positive indirect effect that is conditional on ambidextrous learning; specifically, this indirect effect is least positive for firms with lower ambidextrous learning capacity but stronger for those firms having higher ambidextrous learning capacity. We hold that the key to spurring corporate sustainability performance through green training lies mainly in advancing firms' green innovation behaviors and persistently enhancing their capacities for ambidextrous learning. Thus, this study offers not only new theoretical insights for understanding the effect of green training on corporate sustainability performance but also practical implications for improving corporate sustainability performance within a green economy context. [ABSTRACT FROM AUTHOR]

Spherical-like Tb 3+ and Eu 3+ co-doped Gd 2 O 3 nanoparticles with a particle size around 5.5 nm were synthesized by a polyol route. The optimized luminescence property was obtained when 5 mol% Tb 3+ and 2 mol% Eu 3+ were co-doped. The influence of different polyalcohol solvents (DEG/PEG) on particle size and luminescence properties was investigated. The results show that the nanoparticles Gd 2 O 3 :5%Tb 3+ prepared in PEG presented greater particle size (around 79 nm) and higher luminescence intensity. [ABSTRACT FROM AUTHOR]

Background: Despite growing evidence that N-terminal pro-brain natriuretic peptide (NT-proBNP) has an important prognostic value in older adults, there is limited data on its prognostic predictive value., Objectives: The aim of this study is to evaluate the clinical significance of NT-proBNP in hospitalized patients older than 80 years of age in Beijing, China., Methods: This prospective, observational study was conducted in 724 very elderly patients in a geriatric ward (age ≥80 years, range, 80100 years, mean, 86.6 3.0 years). Multivariate linear regression analysis was used to screen for factors independently associated with NT-proBNP, and the Cox proportional hazard regression model was used to screen for relationships between NT-proBNP levels and major endpoints. The major endpoints assessed were all-cause death and MACEs. P values < 0.05 were considered statistically significant., Results: The prevalence rates of coronary heart disease, hypertension, and diabetes mellitus were 81.4%, 75.1%, and 41.2%, respectively. The mean NT-proBNP level was 770 ± 818 pg/mL. Using multivariate linear regression analyses, correlations were found between plasma NT-proBNP and body mass index, atrial fibrillation, estimated glomerular filtration rate, left atrial diameter, left ventricular ejection fraction, use of betablocker, levels of hemoglobin, plasma albumin, triglycerides, serum creatinine, and blood urea nitrogen. The risk of all-cause death (HR, 1.63; 95% CI, 1.0052.642; P = 0.04) and major adverse cardiovascular events (MACE; HR, 1.77; 95% CI, 1.2893.531; P = 0.04) in the group with the highest NT-proBNP level was significantly higher than that in the group with the lowest level, according to Cox regression models after adjusting for multiple factors. As expected, echocardiography parameters adjusted the prognostic value of NT-proBNP in the model., Conclusions: NT-proBNP was identified as an independent predictor of all-cause death and MACE in hospitalized patients older than 80 years of age.