Search

Unveiling the composition and evolution of ore-forming fluids in porphyry W-Mo mineralization systems is key to understanding their formation processes, but detailed quantitative studies are still limited. The Sansheng porphyry W-Mo deposit (17,285 t WO3 @ 0.569 % and 24,361 t Mo @ 0.226 %), located in southern Great Xing'an Range W belt, NE China, is characterized by quartz vein- and veinlet-disseminated-type W-Mo ore bodies mainly hosted in the cupolas of Early Cretaceous granitic intrusion. In this study, we provide an elaborate study that integrates detailed petrographic, laser Raman, microthermometric, and LA-ICP-MS analyses of fluid inclusions (FIs) to reconstruct the fluid evolution history of the Sansheng W-Mo mineralization system and to bridge the aforementioned knowledge gap. Four stages of hydrothermal veins are identified: stage 1 is the dominant stage of W mineralization, and characterized by the development of wolframite, quartz, K-feldspar, muscovite, and minor amounts of molybdenite; stage 2 is the main stage of Mo mineralization, and featured with the mineral assemblages of molybdenite, quartz, sericite together with minor amounts of pyrite; stage 3 involves Cu-Pb-Zn mineralization, and is characterized by the occurrence of polymetallic sulfides, quartz, and sericite; and stage 4 is dominated by barren carbonate-quartz veins. In wolframite and quartz, the FIs are classified as liquid-rich two-phase (L-type), vapor-rich two-phase (V-type), and pure vapor mono-phase FIs (PV-type) in types. Microthermometric results reveal decreasing homogenization temperatures in primary FIs from stages 1 to 4. Stage 1 fluids are characterized by high homogenization temperatures (320–420 ℃), moderate salinities (7.9–14.8 wt% NaCl equiv.), highly variable W contents (17–7251 ppm) and constant Mo contents (22–288 ppm). FIs in wolframite and coexisting quartz share similar homogenization temperatures and salinities, suggesting that they were formed concurrently. Stage 2 fluids have similar salinities (7.0–13.9 wt% NaCl equiv.) to stage 1 fluids but lower homogenization temperatures (250–340 ℃), and contain highly variable Mo contents (21–11176 ppm) and low W contents (5–23 ppm). Stage 3 fluids display moderate homogenization temperatures (220–300 ℃) and low salinities (3.1–10.1 wt% NaCl equiv.). The contents of W and Mo in stage 3 fluids decrease dramatically, marking the termination of W-Mo mineralization. Stage 4 fluids are characterized by moderate homogenization temperatures (200–270 ℃), low salinities (0.2–5.0 wt% NaCl equiv.), and the lowest contents of W (1–4 ppm) and Mo (8–16 ppm). All stage fluids contain high concentrations of B and Mn, and high ratios of Li/Na, K/Na, Rb/Na, Ba/Na, Pb/Na, and Zn/Na, similar to magmatic-hydrothermal fluids in composition and against typical basinal brines. In addition, they also have high Rb/Sr and low K/Rb ratios, which are equivalent to those of the Sansheng ore-bearing granite. In combination with the constant Cs/(Na + K) and Cs/Rb ratios of all stage FIs, we propose that the multistage hydrothermal fluids originated from a common source–the highly evolved and geochemically uniform Sansheng granitic magma, and that the derivation of ore-forming fluids was a continuous fluid exsolution process. Fluid boiling and fluid-rock interaction are collectively responsible for the deposition of wolframite and molybdenite, while the role of fluid cooling may be subordinate. The meteoric water began to enter the hydrothermal system in stage 3, and fluid cooling and dilution due to the fluid mixing process played a critical role in the deposition of Cu-Pb-Zn sulfides. Our findings highlight the significance of in-situ FI analyses in revealing the composition and evolution of multistage fluids in porphyry W-Mo deposits, which is critical for elucidating detailed ore-forming processes of intrusion-related W-Mo mineralization systems in NE China and globally.

In computer vision, contrastive learning is the most advanced unsupervised learning framework. Yet most previous methods simply apply fixed composition of data augmentations to improve data efficiency, which ignores the changes in their optimal settings over training. Thus, the pre-determined parameters of augmentation operations cannot always fit well with an evolving network during the whole training period, which degrades the quality of the learned representations. In this work, we propose AdDA, which implements a closed-loop feedback structure to a generic contrastive learning network. AdDA works by allowing the network to adaptively adjust the augmentation compositions according to the real-time feedback. This online adjustment helps maintain the dynamic optimal composition and enables the network to acquire more generalizable representations with minimal computational overhead. AdDA achieves competitive results under the common linear protocol on ImageNet-100 classification (+1.11% on MoCo v2)., Accepted by ICASSP 2023 (Oral)

Attention has become a common ingredient in deep learning architectures. It adds a dynamical selection of information on top of the static selection of information supported by weights. In the same way, we can imagine a higher-order informational filter built on top of attention: an Attention Schema (AS), namely, a descriptive and predictive model of attention. In cognitive neuroscience, Attention Schema Theory (AST) supports this idea of distinguishing attention from AS. A strong prediction of this theory is that an agent can use its own AS to also infer the states of other agents' attention and consequently enhance coordination with other agents. As such, multi-agent reinforcement learning would be an ideal setting to experimentally test the validity of AST. We explore different ways in which attention and AS interact with each other. Our preliminary results indicate that agents that implement the AS as a recurrent internal control achieve the best performance. In general, these exploratory experiments suggest that equipping artificial agents with a model of attention can enhance their social intelligence.

Rice bran is a high-quality and renewable livestock feed material rich in nutrients and bioactive substances. To investigate the effects of dietary supplementation with fermented heat-treated rice bran on the performance, apparent digestibility of nutrients, cecal microbiota and metabolites in laying hens, a total of 128 18-week-old Hy-Line brown layers were randomly assigned to four treatment groups: 2.5% HRB (basal diet contained 2.5% heat-treated rice bran), 5.0% HRB (5.0% heat-treated rice bran), 2.5% FHRB (2.5% fermented heat-treated rice bran), 5.0% FHRB (5.0% fermented heat-treated rice bran). Results showed that FHRB supplementation significantly increased the average daily feed intake (ADFI) during 25–28 weeks, and improved apparent digestibility of dry matter (DM), crude protein (CP), ether extract (EE) and crude fiber (CF) in laying hens. Moreover, feeding 5.0% of HRB and FHRB resulted higher egg production (EP) and average egg weight (AEW) during the feeding period, and decreased the feed conversion ratio (FCR) during 21 to 28 weeks. The alpha and beta diversity indices indicated that FHRB altered the cecal microbiota. In particular, dietary supplementation with FHRB significantly increased the relative abundances of Lachnospira and Clostridium. Compared with the 2.5% level of supplementation, supplementing 5.0% HRB and 5.0% FHRB increased the relative abundances of Firmicutes, Ruminococcus and Peptococcus, and lowered the relative abundance of Actinobacteria. Furthermore, dietary FHRB supplementation significantly increased the concentration of short-chain fatty acids in cecum and changed the overall metabolome. The results of correlation analysis showed a close interaction between cecal microbiota, metabolites and apparent digestibility of nutrients. Taken together, we revealed that FHRB supplementation can induce characteristic structural and metabolic changes in the cecal microbiome, which could potentially promote nutrient digestion and absorption, and improve the production performance of laying hens.

ObjectiveThis study aims to examine the association between childhood traumatic events (CTEs), childhood trauma severity, and depressive symptoms, as well as to examine the mediating and moderating roles of resilience in these associations.MethodsWe conducted a cross-sectional study of 1,091 community-dwelling older adults in Jinan, China. The trauma history questionnaire (THQ) was used to measure CTEs and childhood trauma severity. CTEs were defined as the number of traumatic events before the age of 18. We calculated childhood trauma severity by multiplying the number of CTEs by the participants’ self-perceived impact level of the events from the THQ. We then applied the 15-item Geriatric Depression Scale and 10-item Connor–Davidson Resilience Scale to assess participants’ depressive symptoms and resilience, respectively. Linear regression models were used to examine the associations, and structural equation modeling was used to examine the mediating and moderating roles of resilience.ResultsChildhood traumatic events, childhood trauma severity, and resilience were all associated with depressive symptoms in older adults. Resilience mediated the relationship between childhood trauma severity and depressive symptoms (β = 0.082, 95% CI = 0.045–0.123), accounting for 26.6% of the overall effect (β = 0.308, 95% CI = 0.190–0.422). However, there was no evidence that resilience mediated the association between CTEs and depressive symptoms. In addition, we did not find that resilience played a moderating role in the associations of CTEs, childhood trauma severity with depressive symptoms.ConclusionResilience plays a mediating role in the relationship between childhood trauma severity and depressive symptoms. Intervention measures on improving resilience may reduce childhood trauma severity associated with depression risk in older Chinese adults.

The service life of the purging plug is one of the fundamental factors that determine the downtime and usage efficiency of the whole ladle. The creep behaviour of the purging plug was thus investigated to identify the possible failure mechanism. At first, the creep parameters of the Norton–Bailey strain hardening rule were inversely identified via the results of the creep test. Then, the thermal-solid coupling model approach was employed to predict the creep behaviour of the purging plug, in which the Norton–Bailey strain hardening rule was applied. The numerical results show that the temperature of the purging plug presents a cyclic trend after the first service period since the preheating temperature is lower than the temperature of molten steel. Furthermore, the distribution of the creep strain intensity in a layered form could also contribute to a gradual spalling of the purging plug end in service. Besides, the creep strain concentration around the slit can be responsible for the clogging of the purging plug.

The effect of bamboo fiber content on mechanical properties, moisture permeability and light transmittance of composite membrane was studied. The results show that the tensile strength of the composite film is increased by 30%, and the mechanical properties of PVA film are improved obviously with the addition of bamboo cellulose, which can be used as a good reinforcing material of PVA matrix. Bamboo cellulose composite film is a kind of transparent material because of its high light transmittance.

SiC porous ceramics (SPCs) are key functional medium materials, which are widely applied as porous burners. However, SPCs produced via the polyurethane sponge replica technique usually exhibit a randomly distributed cellular structure, leading to vulnerability and a low combustion efficiency for the porous burners. In this study, SPCs were structurally designed via novel 3D printed resin template technique; this endowed the SPCs with synergistically optimized strength and combustion characteristics. SPCs containing tetrakaidecahedron, octahedron and cubic cells were successfully prepared via SiC slurry coating and sintering in air. The cell parameters determined the strength and combustion properties of the SPCs. The SPCs containing cubic cells exhibited the largest strength and strain, because more struts were present parallel to the loading direction; this is more advantageous for stress bearing compared to those vertical to the loading direction. The tetrakaidecahedron cell facilitated the formation of a disturbance effect within the SPCs to a higher extent than the cubic and octahedron cells, thereby enhancing the heat exchange performance between the fluid and porous framework. Hence, the SPC with tetrakaidecahedron cells exhibited the best combustion properties, which resulted in the highest surface temperature and lowest pollution emission.

Understanding how plant pathogenic fungi adapt to their hosts is of critical importance to securing optimal crop productivity. In response to pathogenic attack, plants produce reactive oxygen species (ROS) as part of a multipronged defense response. Pathogens, in turn, have evolved ROS scavenging mechanisms to undermine host defense. Thioredoxins (Trx) are highly conserved oxidoreductase enzymes with a dithiol-disulfide active site, and function as antioxidants to protect cells against free radicals, such as ROS. However, the roles of thioredoxins in Verticillium dahliae, an important vascular pathogen, are not clear. Through proteomics analyses, we identified a putative thioredoxin (VdTrx1) lacking a signal peptide. VdTrx1 was present in the exoproteome of V. dahliae cultured in the presence of host tissues, a finding that suggested that it plays a role in host-pathogen interactions. We constructed a VdTrx1 deletion mutant ΔVdTrx1 that exhibited significantly higher sensitivity to ROS stress, H2O2, and tert-butyl hydroperoxide (t-BOOH). In vivo assays by live-cell imaging and in vitro assays by western blotting revealed that while VdTrx1 lacking the signal peptide can be localized within V. dahliae cells, VdTrx1 can also be secreted unconventionally depending on VdVps36, a member of the ESCRT-II protein complex. The ΔVdTrx1 strain was unable to scavenge host-generated extracellular ROS fully during host invasion. Deletion of VdTrx1 resulted in higher intracellular ROS levels of V. dahliae mycelium, displayed impaired conidial production, and showed significantly reduced virulence on Gossypium hirsutum, and model plants, Arabidopsis thaliana and Nicotiana benthamiana. Thus, we conclude that VdTrx1 acts as a virulence factor in V. dahliae.

In order to solve the problems of single feature dimension and low accuracy in traditional oral English evaluation, this study has used the support vector regression algorithm to achieve the effective integration of different dimension evaluation features. The study comprehensively evaluate the overall reading and pronunciation quality of students. This paper designs an evaluation model of spoken English pronunciation quality based on a dynamic time warping algorithm and introduces the evaluation algorithms of pronunciation standard, fluency, and intonation, respectively. A support vector regression algorithm is used to realize the effective fusion of different dimension evaluation features. In addition, the correlation between machine rating and manual rating was used as an evaluation index to evaluate students' oral pronunciation quality from different dimensions. The model shows superior performance in the pronunciation of both monosyllabic and polysyllabic words, which is beneficial to improving students' oral English learning effect.

We present a novel multimodal interpretable VQA model that can answer the question more accurately and generate diverse explanations. Although researchers have proposed several methods that can generate human-readable and fine-grained natural language sentences to explain a model's decision, these methods have focused solely on the information in the image. Ideally, the model should refer to various information inside and outside the image to correctly generate explanations, just as we use background knowledge daily. The proposed method incorporates information from outside knowledge and multiple image captions to increase the diversity of information available to the model. The contribution of this paper is to construct an interpretable visual question answering model using multimodal inputs to improve the rationality of generated results. Experimental results show that our model can outperform state-of-the-art methods regarding answer accuracy and explanation rationality., Under review

Despite advances in cervical cancer screening and human papilloma virus (HPV) vaccines, cervical cancer remains a global health burden. The standard treatment of cervical cancer includes surgery, radiation therapy, and chemotherapy. Radiotherapy (RT) is the primary treatment for advanced-stage disease. However, due to radioresistance, most patients in the advanced stage have an adverse outcome. Recent studies have shown that long noncoding RNAs (lncRNAs) participate in the regulation of cancer radiosensitivity by regulating DNA damage repair, apoptosis, cancer stem cells (CSCs), and epithelial–mesenchymal transition (EMT). In this review, we summarize the molecular mechanisms of long noncoding RNAs in cervical cancer and radiosensitivity, hoping to provide a theoretical basis and a new molecular target for the cervical cancer RT in the clinic.

Understanding the impact of drought on fire dynamics is crucial for assessing the potential effects of climate change on wildfire activity in China. In this study, we present a series of multiple linear regression (MLR) models linking burned area (BA) during mainland China’s fire season from 2001 to 2019, across seven regions, to concurrent drought, antecedent drought, and time trend. We estimated burned area using Collection 6 Moderate Resolution Imaging Spectradiometer (MODIS) and drought indicators using either the Standardized Precipitation Evapotranspiration Index (SPEI) or the self-calibrated Palmer Drought Severity Index (sc-PDSI). Our findings indicate that the wildfire season displays a spatial variation pattern that increases with latitude, with the Northeast China (NEC), North China (NC), and Central China (CC) regions identified as the primary areas of wildfire occurrence. Concurrent and antecedent drought conditions were found to have varying effects across regions, with concurrent drought as the dominant predictor for NEC and Southeast China (SEC) regions and antecedent drought as the key predictor for most regions. We also found that the Northwest China (NWC) and CC regions exhibit a gradual decrease in burned area over time, while the NEC region showed a slight increase. Our multiple linear regression models exhibited a notable level of predictive power, as evidenced by the average correlation coefficient of 0.63 between the leave-one-out cross-validation predictions and observed values. In particular, the NEC, NWC, and CC regions demonstrated strong correlations of 0.88, 0.80, and 0.76, respectively. This indicates the potential of our models to contribute to the prediction of future wildfire occurrences and the development of effective wildfire management and prevention strategies. Nevertheless, the intricate relationship among fire, climate change, human activities, and vegetation distribution may limit the generalizability of these findings to other conditions. Consequently, future research should consider a broad range of factors to develop more comprehensive models.

Hierarchical text classification (HTC) is a challenging subtask of multi-label classification as the labels form a complex hierarchical structure. Existing dual-encoder methods in HTC achieve weak performance gains with huge memory overheads and their structure encoders heavily rely on domain knowledge. Under such observation, we tend to investigate the feasibility of a memory-friendly model with strong generalization capability that could boost the performance of HTC without prior statistics or label semantics. In this paper, we propose Hierarchy-aware Tree Isomorphism Network (HiTIN) to enhance the text representations with only syntactic information of the label hierarchy. Specifically, we convert the label hierarchy into an unweighted tree structure, termed coding tree, with the guidance of structural entropy. Then we design a structure encoder to incorporate hierarchy-aware information in the coding tree into text representations. Besides the text encoder, HiTIN only contains a few multi-layer perceptions and linear transformations, which greatly saves memory. We conduct experiments on three commonly used datasets and the results demonstrate that HiTIN could achieve better test performance and less memory consumption than state-of-the-art (SOTA) methods., Comment: Accepted by ACL'23

Recently, more and more people study online for the convenience of access to massive learning materials (e.g. test questions/notes), thus accurately understanding learning materials became a crucial issue, which is essential for many educational applications. Previous studies focus on using language models to represent the question data. However, test questions (TQ) are usually heterogeneous and multi-modal, e.g., some of them may only contain text, while others half contain images with information beyond their literal description. In this context, both supervised and unsupervised methods are difficult to learn a fused representation of questions. Meanwhile, this problem cannot be solved by conventional methods such as image caption, as the images may contain information complementary rather than duplicate to the text. In this paper, we first improve previous text-only representation with a two-stage unsupervised instance level contrastive based pre-training method (MCL: Mixture Unsupervised Contrastive Learning). Then, TQ-Net was proposed to fuse the content of images to the representation of heterogeneous data. Finally, supervised contrastive learning was conducted on relevance prediction-related downstream tasks, which helped the model to learn the representation of questions effectively. We conducted extensive experiments on question-based tasks on large-scale, real-world datasets, which demonstrated the effectiveness of TQ-Net and improve the precision of downstream applications (e.g. similar questions +2.02% and knowledge point prediction +7.20%). Our code will be available, and we will open-source a subset of our data to promote the development of relative studies., Comment: This paper has been accepted for the AAAI2023 AI4Edu Workshop

In this article, room temperature gallium-based fluids (RTGFs) with unique thermal and conductive properties are proposed as a transmission fluid for the force carrying medium of hydraulics, which rectify the lack of the multi-functional hydraulic roles of liquid metal (LM). The typical physical properties of RTGFs and comparative conventional hydraulic fluids (commercial hydraulic oil and deionized water), such as thermal stability and rheological characteristics are evaluated. Experimental and numerical methods, then, are adopted to clarify the force transmission performance of RTGFs and commercial hydraulic oils, as well as the influence of temperature fields on the viscosity of fluids. The results disclosed that the advantages of inherent flame resistance, wide liquid temperature range, and the viscosity changing slightly with temperature, make RTGFs potential to efficiently apply in the hydraulics as new working fluids. Finally, for illustration, the rigid and flexible actuators driven by RTGFs were designed as hydraulic fluid and demonstrated their capabilities in grasping objects with various shapes and weights, respectively. And the tunable stiffness of such a flexible actuator is enabled via the solid-liquid phase change of LM. Additionally, a frequency-adjustable antenna was manufactured and showcased owing to the introduced transformable electromagnetic behaviors of LM. Overall, the gallium-based LM fluids with thermoelectrical and mechanical multimodal signal medium, would serve as a potential candidate for future complex multifunctional signal transmission systems.

It plays crucial importance for practical application to clarify the coupled nonlinear effects of multi-parameter uncertainty from temperature-dependent material properties, electrical and thermal working conditions, configuration, and geometry size on the performance of thermoelectric generator (TEG). This paper develops a comprehensive tool for optimal geometry design of the multi-stage TEG and identifies the synergistic impacts of the multi-parameter uncertainty on its optimal performance. A novel balanced impedance matching model is established for the first time to balance the optimal output power and conversion efficiency. The developed method is then efficiently integrated with optimization and uncertainty analysis methods. The designed three-stage TEG could achieve better performance due to the optimal working temperature range obtained for each stage, reaching an optimal efficiency of 18.06% under a temperature difference of 500 K. Compared to the single-stage TEG, a remarkable increase of 36.40% for output power and 34.47% for conversion efficiency are achieved for the convective heat transfer conditions, respectively. Uncertainty analysis has indicated that the uncertainty of Seebeck coefficients, leg length ratios, and working temperature conditions have significant impacts on the performance of multi-stage TEG. In contrast, the uncertainty of leg cross-section area ratios and external electrical loading have less effect. The coupled effects of multiple parameters with a total number up to 26 for three-stage TEG are studied in detail. Overall, this work provides a promising comprehensive tool for intensifying TEG performance and clarifying the nonlinear coupled effects of multi-parameter inherent uncertainty.

Human intelligence is able to first learn some basic skills for solving basic problems and then assemble such basic skills into complex skills for solving complex or new problems. For example, the basic skills "dig hole," "put tree," "backfill" and "watering" compose a complex skill "plant a tree". Besides, some basic skills can be reused for solving other problems. For example, the basic skill "dig hole" not only can be used for planting a tree, but also can be used for mining treasures, building a drain, or landfilling. The ability to learn basic skills and reuse them for various tasks is very important for humans because it helps to avoid learning too many skills for solving each individual task, and makes it possible to solve a compositional number of tasks by learning just a few number of basic skills, which saves a considerable amount of memory and computation in the human brain. We believe that machine intelligence should also capture the ability of learning basic skills and reusing them by composing into complex skills. In computer science language, each basic skill is a "module", which is a reusable network of a concrete meaning and performs a specific basic operation. The modules are assembled into a bigger "model" for doing a more complex task. The assembling procedure is adaptive to the input or task, i.e., for a given task, the modules should be assembled into the best model for solving the task. As a result, different inputs or tasks could have different assembled models, which enables Auto-Assembling AI (AAAI). In this work, we propose Modularized Adaptive Neural Architecture Search (MANAS) to demonstrate the above idea. Experiments on different datasets show that the adaptive architecture assembled by MANAS outperforms static global architectures. Further experiments and empirical analysis provide insights to the effectiveness of MANAS., In ACM CIKM 2022

Purging plug with slits is widely used in the secondary refining process, and its service life determines the downtime and production efficiency. To achieve a good performance in service life, optimizing the geometric structure of the slit, which can alleviate the stress concentration around the slit, was employed using the finite plate model. Furthermore, the performance of the purging plugs with different shapes of slits in the refining effect was investigated by conducting the water model experiments. The numerical simulation results showed that the appropriate shapes of slits could alleviate the stress concentration phenomena. In the water model experiment, the mixing time of the purging plug with circular slits is lesser than that of the purging plug with rectangular slits when the gas flow is less than 6.02 L/min. Compared with the rectangular slits, the circular slits are detrimental to smaller and well-distributed bubbles. Thus, the inclusion removal rate of the purging plug with circular slits is approximately 10% larger than that of the purging plug with the rectangular slits.

Purging plugs employed in ladles in the steel industry undergo severe thermomechanical loads owing to the cyclic operations, including preheating, transporting, stirring, and holding stages. For the smooth running of the process with less downtime, the lifespan of the purging plug needs to be aligned with the overhaul interval without failure. In the present study, to identify the thermomechanical failure mechanisms of corundum castable purging plugs with different slit structures via a quantitative evaluation method, the Drucker–Prager material constitutive model was applied to account for the irreversible behavior. The results indicate that both shear and tensile failure occur in the purging plug with rectangular slits, where the shear failure dominates in the middle of the holding stage and tensile failure appears during the stirring stage. Moreover, the premature shear failure occurs in the upper section near the working face for both types of purging plugs. The application of circular slits diminishes the stress concentration in the purging plug and changes the fracture mechanisms, extent, and occurrence time, which experience mere shear failure in the early stage of transportation.

The single-lap joints with different overlap lengths were bonded by carbon fiber reinforced plastics (CFRP) laminate, and their failure load, fracture process and strain fields were characterized by digital image correlation (DIC) and universal testing machine. The effects of overlap length on the tensile properties, fracture process, strain distribution and failure characteristics of single lap joints were studied. The results show that with the development of overlap length, the average shear strength of joints is decreased firstly and then tends to be stable. The secondary bending effect of joints caused by the eccentric load during the stretching process becomes more significant with the increase of overlap length, the end deformation of the lap area is increased, and the initial failure position of joints changes from one end to both ends of the lap area. The strain concentration area at the end of the front and side of joints transitions from asymmetric to symmetric. The peeling stress of joints gradually is increased, and the main failure mode between layers is changed from shear to peeling failure. The failure mode of joints go through the process from interface and slight fiber tearing to mixed failure and then to delamination.

BackgroundSurgical resection combined with oral tyrosine kinase inhibitors(TKI) is the most effective treatment for gastrointestinal stromal tumor(GIST) liver metastases. Liver transplantation (LT) is the last resort for the treatment of terminal liver malignancy. Whether it can be a potential treatment option for liver metastases from unresectable GIST is worth exploring.Case presentationWe report a 38-year-old woman who underwent jejunal stromal tumor resection and TKI(imatinib) therapy 15 years ago for jejunal stromal tumor liver metastases. During the period from 2015 to 2018, the liver metastases continued to grow after the patient stopped taking imatinib voluntarily, and the patient subsequently underwent multiple interventional surgeries and drug treatments, which were still poorly curative. The tumor was deemed unresectable because it filled the entire liver, and the patient subsequently underwent LT and was treated with imatinib post-operatively, which resulted in no recurrence of the tumor within 18 months of follow-up.Literature reviewThere are few reports in the literature on LT for the treatment of liver metastases from GIST. A systematic review and summary of the current literature by literature search revealed that LT as a last resort for metastatic GIST of the liver remains a major challenge.ConclusionsLT combined with TKI-targeted therapy is a potential therapy worth exploring for patients with unresectable metastatic GIST.

A water model experiment and a transient 3D numerical model were designed and implemented to quantitatively evaluate the flow-induced erosion behavior of the refractory lining in a continuous casting tundish. A relationship between the erosion rate and flow characteristic parameters was proposed and experimentally verified. Boric acid tablets were used in the water model experiment to determine the flow-induced erosion rate, and the numerical model defined the two-phase flow pattern. The results indicate that the impacting and rubbing effects caused by the fluid flow can be quantitatively characterized by the turbulent kinetic energy and wall shear stress. The proposed empirical dependence describes the erosion rate of the boric acid tablet with acceptable accuracy. The most serious erosion occurs inside the turbulence inhibitor, and the erosion rate sharply drops and steadily reduces with the water flow outside the turbulence inhibitor. The influence of casting speed on the refractory erosion rate is also quantitatively analyzed using the developed model with the introduction of the proposed correction.

Channel attention mechanisms have attracted more and more researchers because of their generality and effectiveness in deep convolutional neural networks(DCNNs). However, the signal encoding methods of the current popular channel attention mechanisms are limited. For example, SENet uses the full-connection method to encode channel relevance, which is parameters-costly; ECANet uses 1D-Convolution to encode channel relevance, which is parameter fewer but can only encode per k adjacent channels in a fixed scale. This paper proposes a novel dilated efficient channel attention module(DECA), which consists of a novel multi-scale channel encoding method and a novel channel relevance feature fusion method. We empirically show that different scale channel relevance also contributes to performance, and fusing various scale channel relevance features can obtain more powerful channel feature representation. Besides, we widely use the weight-sharing method in the DECA module to make it more efficient. Specifically, we have applied our module to the real-life fire image detection task to evaluate its effectiveness. Extensive experiments on different backbone depths, detectors, and fire datasets have shown that the average performance boost of DECA module is more than 4.5% compare to the baselines. Meanwhile, DECA outperforms other state-of-art attention modules while keeping lower or comparable parameters in the experiments. The experimental results on different datasets also shown that the DECA module holds great generalization ability.

The new route for recycling fly ash was proposed to prepare SiC reticulated porous ceramics (SRPCs) with high strength and increased efficient filtration for molten metal filtration. The effects of fly ash on the rheological characteristics, microstructure evaluation and wetting behavior between SRPCs and molten metal were investigated. It was found that the fly ash was beneficial to thixotropic property of SiC slurry when its content was less than 30 wt%. Furthermore, fly ash in SRPCs was completely transformed into mullite with needle-shape at 1300 °C, forming a porous structure containing micro pores and windows. SRPCs containing 20 wt% fly ash exhibited a higher strength because of the improved rheological properties of SiC slurry and the optimized microstructure in skeleton. In addition, the added fly ash in SRPCs could increase the contact angle between skeleton substrate and molten metal via microporosization of skeleton, thus exhibiting the potential ability to improve the filtration efficiency.

The formation and removal of exogenous inclusions in a real-size two-strand tundish is simulated by the proposed unsteady 3D comprehensive numerical model of the respective fluid-structure interaction, which takes into account the impacting and washing effects on the refractory wear. A large eddy simulation is employed to describe the molten steel vortex flow. Thus, the thermal profiles of the molten steel and refractory lining are constructed. One-way coupled unsteady Euler-Lagrange approach is adopted to estimate the detachment and motion of the exogenous inclusion. The inclusion’s Reynolds number is utilized for evaluating the inclusion separation at the refractory lining after formation and at the upper surface of the molten steel. At a 1.2 m/min casting speed, 49 and 38 pct of exogenous inclusions are created at the turbulent inhibitor inner bottom and long nozzle inner wall, respectively. In contrast, only 13 pct of new inclusions are produced at all other inner walls. About 80 pct of newly generated inclusions are then trapped by free surfaces, 78 pct of which are removed at the first free surface. The initial diameter of exogenous inclusions ranges from 13 to 48 μm. The removal ratio of exogenous inclusions in the tundish first grows from 61 to 80 pct, with the casting speed rising from 1.0 to 1.2 m/min and then drops to 63 pct after the further casting speed rise to 1.4 m/min.

The hot metal desulfurization in Kanbara Reactor (KR) metal treatment was simulated in this study via a transient-coupled 3D numerical model of the two-phase flow, heat transfer, and particle motion desulfurization processes. The KR impeller stirring was described via the multiple reference frame model. The volume of fluid approach was employed to capture the air-hot metal interface. The particle motion and aggregation were defined by the two-way coupled Euler–Lagrangian method. A desulfurization kinetic model was simultaneously introduced to represent the sulfur mass transfer rate. The effect of the initial diameter of desulfurizing agent (DA) particles on the desulfurization efficiency was quantitatively assessed. The lowest sulfur content was observed in the impeller vicinity and the highest one in the inactive colder liquid metal at the vessel bottom. With the DA particle initial diameter reduction from 3.0 to 0.5 mm, the overall desulfurization rate was increased from 83.2 to 97.1 pct. Insofar the specific surface area of smaller particles exceeded that of larger ones, they had higher motion velocities and heating rates, creating a greater reactivity for the desulfurization. However, desulfurization at the vessel bottom was only slightly enhanced using smaller DA particles, so the overall improvement did not exceed 20 pct. Further enhancement is envisaged by refining the impeller design and particle-adding technique.

Electroslag remelting (ESR) technology has been adopted to recycle the rejected electrolytic manganese metal scrap. In this study, a transient 3D coupled numerical model accounting for the electromagnetism, multiphase flow, heat transfer, and solidification was elaborated and used to simulate the evaporation behavior of the molten manganese metal (MM) during the ESR process. The volume of fluid approach was employed to capture the interfaces between the gaseous manganese, molten slag, and molten MM. The evaporation rate of the molten MM was defined by applying the Lee model, while the enthalpy-porosity formulation described the solidification. An industrial experiment via a commercial-scale ESR furnace was conducted for the model validation. The research findings indicate that the molten MM’s evaporation occurs during the droplet falling and in the metal pool. Then, gaseous manganese bubbles ascend to the molten slag-free surface, thus promoting the melt movement, especially the slag-metal pool interface fluctuation. The evaporation rate of the molten MM is promoted by the increased applied current and the reduced ambient gauge pressure. The recycling ratio drops from 81.75 to 71.79 pct with the applied current increase from 3000 to 4000 A and drops from 78.19 to 73.71 pct with the ambient gauge pressure reduction from 0 to − 1000 Pa.

Reusable electronics have received widespread attention and are urgently needed. Here, nanocellulose-based liquid metal (NC-LM) printed circuit has been fabricated by the evaporation-induced transfer printing technology. In this way, the liquid metal pattern is embedded into the nanocellulose membrane, which is beneficial for the stability of the circuit during use. Besides, the NC-LM circuit is ultrathin with just tens of microns. In particular, the finished product is environmentally friendly because it can be completely dissolved by water, and both the liquid metal ink and the nanocellulose membrane can be easily recollected and reused, thereby reducing waste and pollution to the environment. Several examples of flexible circuits have been designed to evaluate their performance. The mechanism of evaporation-induced transfer printing technology involves the deposition, aggregation, and coverage tightly of the nanosized cellulose fibrils as the water evaporated. This study provides an economical and environmentally friendly way for the fabrication of renewable flexible electronics.

Rice paddy is a significant source of atmospheric methane (CH

During the last decades, China had issued a series of stringent control measures, resulting in a large decline in air pollutant concentrations. To quantify the net change in air pollutant concentrations driven by emissions, we developed an approach of determining the closed interval of the deweathered percentage change (DPC) in the concentration of air pollutants on an annual scale, as well as the closed intervals of cumulative DPC in a year compared with that in the base year. Thus, the hourly mean mass concentrations of criteria air pollutants to determine their interannual variations and the closed intervals of their DPCs during the heating seasons from 2013 to 2019 in Qingdao (a coastal megacity) were analyzed. The seasonal mean SO2 concentration decreased from 2013 to 2019. The seasonal mean CO, NO2, and PM2.5 concentrations also generally decreased from 2013 to 2017, but increased unexpectedly in 2018 (from 0.9 mg m−3 (CO), 42 µg m−3 (NO2), and 51 µg m−3 (PM2.5) in 2017 to 1.1 mg m−3, 48 µg m−3, and 64 µg m−3 in 2018, respectively). The closed intervals of DPC in concentrations of CO, NO2, and PM2.5 from the 2017 heating season (2017/2018) to the 2018 heating season (2018/2019) were obtained at (27%, 30%), (15%, 18%), and (30%, 33%), respectively. Such high positive endpoint values of the closed intervals, in contrast to their small interval lengths, indicate increased emissions of these pollutants and/or their precursors in 2018/2019 compared with 2017/2018, by minimizing the meteorological influences. The rebounds of CO, NO2, and PM2.5 in 2018/2019 were likely associated with a doubled increase in natural gas (NG) consumption implemented by the “coal-to-NG” project, as the total energy consumption showed little difference. Our results suggested an important role of the “coal-to-NG” project in driving concentrations of air pollutant increases in China in 2018/2019, which need integrated assessments.

Background Surgical resection combined with oral tyrosine kinase inhibitors(TKI) is the most effective treatment for gastrointestinal stromal tumor(GIST) liver metastases. For patients with TKI-resistant and unresectable liver metastatic GIST, there has been a lack of effective treatment measures. Liver transplantation is the last resort for end-stage liver malignancies, and it is worth exploring as a treatment option for GIST liver metastases. Case presentation: We report a 38-year-old woman who underwent jejunal stromal tumor resection and TKI(imatinib) therapy 15 years ago for jejunal stromal tumor liver metastases. During the period from 2015 to 2018, the liver metastases continued to grow after the patient stopped taking imatinib voluntarily, and the patient subsequently underwent multiple interventional surgeries and drug treatments, which were still poorly curative. The tumor was deemed unresectable because it filled the entire liver, and the patient subsequently underwent liver transplantation and was treated with imatinib post-operatively, which resulted in no recurrence of the tumor within 13 months of follow-up. Conclusions Liver transplantation can be used as a treatment option for patients with unresectable liver metastatic GIST, and combined TKI-targeted therapy can prolong survival to a certain extent.

Two types of first-order decoupled conservative schemes are firstly proposed for the strongly coupled nonlinear Schrodinger system by using pseudospectral method in space and coordinate increment discrete gradient method in time. And then, in order to improve the solution accuracy in time, the composition methods are employed to construct second- and fourth-order schemes. The proposed schemes are efficient for the system in d ≥ 2 dimensions and also easy to code because of their decoupled feature. A fast solver is proposed to speed up the computation. Ample numerical examples including the motion of single soliton and interaction of multiple solitary waves are carried out to exhibit the performance of the schemes.

Xue Yan,* Dehao Zeng,* He Zhu, Yijing Zhang, Yuying Shi, Yingxiu Wu, Hongmei Tang, Detang Li Pharmaceutical Department, The First Affiliated Hospital of Guangzhou University of Chinese Medicine, Guangzhou, Guangdong Province 510405, People’s Republic of China*These authors contributed equally to this workCorrespondence: Xue Yan Tel +86-020-36598864Email zd0171@163.comBackground: It is known that miR-532-5p is critical for neuronal differentiation. However, the role of miR-532-5p in depression remains unknown. This study aimed to investigate the role and mechanism of miR-532-5p in major depressive disorder (MDD).Methods: The depression mice model was established by chronic unpredictable mild stress (CUMS) and confirmed by forced swimming test (FST) and sucrose preference test (SPT). The role of miR-532-5p in MDD was detected by tail suspension test (TST), FST, SPT and SIT. QRT-PCR was used to detect the expression of miR-139-5p in hippocampus and BV-2 microglia of mice. ELISA and Western blotting were used to detect the expression of the nitric oxide synthase (NOS), proinflammatory cytokines (IL-6, IL-1β, TNF-α, and MCP-1) and transcriptional activator 3 (STAT3). Luciferase reporter assay was used to verify the downstream target genes of miR-532-5p.Results: MiR-532-5p was significantly reduced in the hippocampus of mice treated with CUMS. Overexpression of miR-532-5p significantly reduced CUMS-induced depression-like behaviors and suppressed the expression of IL-6, IL-1β, TNF-α and MCP-1. MiR-532-5p directly targeted signal transducers and STAT3 in BV2 cells. In addition, overexpression of miR-532-5p restrained the raise of inducible NOS and IL-6, IL-1 β, TNF-α and MCP-1 in LPS-exposed BV2 cells.Conclusion: This study indicates that miR-532-5p plays an important role in CUMS-induced depression-like behaviors by targeting STAT3, and miR-532-5p may be a potential target for MDD therapy.Keywords: miR-532-5p, STAT3, major depressive disorder, inflammation