A uniform statistical inferential tool in making individualized treatment decisions, which implements the methods of Ma et al. (2017)<doi:10.1177/0962280214541724> and Guo et al. (2021)<doi:10.1080/01621459.2020.1865167>. It uses a flexible semiparametric modeling strategy for heterogeneous treatment effect estimation in high-dimensional settings and can gave valid confidence bands. Based on it, one can find the subgroups of patients that benefit from each treatment, thereby making individualized treatment selection.
The purpose of this study was to examine the daily social pressure and socioeconomic factors related to women’s alcohol consumption in China. Cross-sectional data were obtained from the 2012 China Family Panel Studies. A multivariate logistic regression analysis of a sample of 16 339 female adults with the mean age of 45.3 years was used to examine the relationships between dependent and independent variables. According to the results, first, the greater the daily social pressure, the more likely women were to engage in general alcohol consumption (odds ratio = 1.061) and risk drinking (odds ratio = 1.057). Second, while there is a positive relationship between the general level of social pressure and women’s alcohol consumption, the relationship between the severe level of social pressure and women’s alcohol consumption was not significant. Finally, women in the Central region were less likely to engage in risk drinking than women in the Western region; women with secondary school education were more likely to engage in risk drinking than women with primary school education or below; and age was significantly positively associated with both general and risk drinking. In conclusion, increasing alcohol consumption among women may be due to increased social pressure.
Deep Spiking Neural Networks (SNNs) present optimization difficulties for gradient-based approaches due to discrete binary activation and complex spatial-temporal dynamics. Considering the huge success of ResNet in deep learning, it would be natural to train deep SNNs with residual learning. Previous Spiking ResNet mimics the standard residual block in ANNs and simply replaces ReLU activation layers with spiking neurons, which suffers the degradation problem and can hardly implement residual learning. In this paper, we propose the spike-element-wise (SEW) ResNet to realize residual learning in deep SNNs. We prove that the SEW ResNet can easily implement identity mapping and overcome the vanishing/exploding gradient problems of Spiking ResNet. We evaluate our SEW ResNet on ImageNet, DVS Gesture, and CIFAR10-DVS datasets, and show that SEW ResNet outperforms the state-of-the-art directly trained SNNs in both accuracy and time-steps. Moreover, SEW ResNet can achieve higher performance by simply adding more layers, providing a simple method to train deep SNNs. To our best knowledge, this is the first time that directly training deep SNNs with more than 100 layers becomes possible. Our codes are available at https://github.com/fangwei123456/Spike-Element-Wise-ResNet.
Pharmaceuticals and personal care products (PPCPs) are of great concern due to their increasing health effects, so advanced treatment technologies for PPCPs removal are urgently needed. In this study, titanate nanotubes decorated Co(OH)2 hollow microsphere (CoM/TNTs) composites were synthesized by a two-step solvothermal method, and used to activate peroxymonosulfate (PMS) through heterogenous catalysis for acetaminophen (ACE) degradation in water. The optimum material (CoM/TNTs0.5) activated PMS system exhibited high ACE removal efficiency and quick kinetic, as 93.0% ACE was degraded even within 10 min. The two components in CoM/TNTs showed a synergetic effect on PMS activation for radicals production: Co(OH)+ from CoM was the primary active species to active PMS, while TNTs could offer abundant –OH groups for Co(OH)+ formation. Density functional theory (DFT) calculation further interpreted the mechanism of Co(OH)+ for PMS activation by means of reaction potential energy surface (PES) analysis. Both the scavenger quenching tests and electron paramagnetic resonance analysis revealed that the sulfate radical (SO4-·) played a dominant role in ACE degradation. Moreover, DFT calculation also suggested that the ACE atoms with high Fukui index (f -) represented the active sites for electrophilic attack by SO4-·. The toxicity analysis based on quantitative structure-activity relationship (QSAR) verified the reduced toxicity of transformation products. Furthermore, CoM/TNTs also had good reusability and stability over five cycles. This work provides deep insights into the reaction mechanisms of radical production and organics attack in cobalt-based PMS activation system.
Degradation pathway is important for the study of carbamazepine (CBZ) removal in advanced oxidation processes (AOPs). Generally, degradation pathways are speculated based on intermediate identification and basic chemical rules. However, this semiempirical strategy is sometimes time-consuming and baseless. To improve the situation, a mini meta-analysis was first conducted for the degradation pathways of CBZ in AOPs. Then, the rationality of the pathways was analyzed by Density Functional Theory (DFT) calculation. Results show that the degradation pathways of CBZ in various AOPs has high similarity, and the reactive sites predicted by Fukui function fitted well with the data retrieved from literatures. In addition, molecule configuration of degradation intermediates was found to play a very important roles on degradation pathway. The study reveals that computational chemistry is a useful tool for degradation pathway speculation in AOPs.