Search

OBJECTIVE The objective of this study was to investigate the incidence and predictors of first-pass effect (FPE) and to compare the clinical outcomes among FPE, multiple-pass effect, and incomplete reperfusion (ICR) in acute anterior circulation large vessel occlusion. METHODS Patients from the ANGEL-ACT (Endovascular Treatment Key Technique and Emergency Workflow Improvement of Acute Ischemic Stroke) registry were included. FPE was defined as complete reperfusion after a single pass of the thrombectomy device without rescue treatment. MPE was defined as complete reperfusion after ≥ 2 passes of the thrombectomy device. ICR was defined as a modified Thrombolysis in Cerebral Infarction score of 2b independent of the number of passes. Multivariable analyses were used to determine predictors of FPE and to compare the following outcomes: functional independence, mortality within 90 days, intraprocedural complications, and intracranial hemorrhage (ICH) among FPE, MPE, and ICR. RESULTS There were 1139 patients, of whom 307 (27.0%) achieved FPE. FPE was related to occlusion location (M1 vs internal carotid artery [ICA], adjusted OR [aOR] 1.57, 95% CI 1.15–2.15, p = 0.004; M2/anterior cerebral artery vs ICA, aOR 2.06, 95% CI 1.32–3.22, p = 0.002) and negatively associated with underlying intracranial atherosclerosis disease (ICAD) (aOR 0.33, 95% CI 0.23–0.49, p < 0.001). Patients with FPE had a higher rate of functional independence (52.7% of FPE patients vs 45.6% of MPE patients and 37.1% of ICR patients, p = 0.002; MPE vs FPE, aOR 0.69, 95% CI 0.51–0.95, p = 0.023; ICR vs FPE, aOR 0.45, 95% CI 0.31–0.66, p < 0.001), lower rate of intraprocedural complications (4.2% vs 18.1% and 21.2%, p < 0.001; MPE vs FPE, aOR 6.23, 95% CI 3.36–11.54, p < 0.001; ICR vs FPE, aOR 7.70, 95% CI 3.97–14.94, p < 0.001), and lower rate of ICH within 24 hours (18.3% vs 27.9% and 26.9%, p = 0.009; MPE vs FPE, aOR 1.97, 95% CI 1.35–2.86, p < 0.001; ICR vs FPE, aOR 2.03, 95% CI 1.30–3.16, p = 0.002) than those with MPE and ICR. CONCLUSIONS FPE was achieved at a rate of 27.0% and associated with functional independence, decreased intraprocedural complications, and ICH. Non-ICA occlusion and underlying ICAD were predictors of FPE.

Background: Sex disparities in acute large vessel occlusion (LVO) following endovascular treatment (EVT) have been recently reported. However, there is uncertainty about the effect of sex differences on functional outcomes after EVT, particularly in an Asian population. The present study aimed to compare the clinical and safety outcomes between men and women with anterior circulation LVO treated with EVT. Methods: We analyzed data from the ANGEL-ACT (Endovascular Treatment Key Technique and Emergency Work Flow Improvement of Acute Ischemic Stroke: a Prospective Multicenter Registry Study) Registry, which was conducted at 111 hospitals from 26 provinces in China between November 2017 and March 2019. Men and women with anterior circulation LVO treated with EVT were matched using propensity scores. After a 1:1 propensity score matching, we compared the clinical outcomes including 90-day ordinal modified Rankin Scale distribution (primary outcome), procedure duration, successful reperfusion, symptomatic intracranial hemorrhage, and mortality. Furthermore, we explored sex modification on the primary outcome in subgroup analysis. Results: Of 1321 patients, 483 (36.6%) were women and 838 (63.4%) were men. The mean age for women and men were 68 and 62 years old, respectively. Among 578 patients identified after matching, there were no sex differences (men versus women) in 90-day ordinal modified Rankin Scale distribution (median [interquartile range], 4 [1–5] versus 3 [1–5], P =0.464), successful reperfusion (86.5% versus 91.0%, P =0.089), symptomatic intracranial hemorrhage (6.5% versus 7.9%, P =0.512), and mortality within 90 days (17.7% versus 17.0%, P =0.826). However, men had a longer median procedure duration than women (86 [52–128] versus 72 [48–110] minutes, β=14.51, [95% CI, 4.19–24.84]; P =0.006). Subgroup analysis showed that in patients with National Institutes of Health Stroke Scale score P for interaction=0.032). Conclusions: Overall, this matched-control study from the ANGEL-ACT study showed similar clinical outcomes between men and women with anterior circulation LVO treated with EVT. However, in the subgroup of patients presenting with lower stroke severity (ie, National Institutes of Health Stroke Scale score Registration: URL: https://www.clinicaltrials.gov ; Unique identifier: NCT03370939.

From the perspective of liquor brewing technology, the quality of liquor yeast undoubtedly determines the quality of liquor products, but the problems such as inconvenient storage, difficult transportation and easy deterioration of liquid liquor yeast greatly restrict the development of liquor industry. Aiming at this problem, the author firstly summarized the research status of drying technology, damage mechanism and protection strategy of white spirit yeast. Then, on the basis of studying the damage mechanism of yeast in the drying process of yeast, the optimization of drying process and the formulation of protective strategies of yeast were discussed. Finally, new research methods are proposed from three perspectives: optimal design of drying process, damage mechanism and protection strategy

A subgroup of patients with acute large vessel occlusion (ALVO) may experience delayed neurological improvement (DNI) after endovascular treatment (EVT). Our study aimed to investigate the incidence and independent predictors of DNI in patients with ALVO after EVT.We selected subjects from ANGEL-ACT Registry. The definition of DNI is patients with ALVO who did not experience early neurological improvement (ENI) despite complete recanalization after EVT. These patients achieved a 90-day favorable outcome assessed by a modified Rankin Scale (mRS) score. We defined ENI as a ≥ 4-point decrease in the National Institutes of Health Stroke Scale (NIHSS) between baseline and 24 h or NIHSS of 0 or 1 at 24 h, with complete recanalization after EVT. We performed logistic regression analyses to determine the independent predictors of DNI.Among the 1056 enrolled patients, 406 (38.4%) did not experience ENI. 106 (26.1%) patients without ENI achieved DNI. On Multivariate analysis, lower admission NIHSS score (odds ratio [OR] = 1.17,95% confidence interval [CI]: 1.11-1.23, P 0.001), underlying ICAD (OR = 2.03, 95% CI: 1.07-3.85, P = 0.029) and absence of general anesthesia (OR = 2.13, 95% CI: 1.24-3.64, P = 0.006) were independent predictors of DNI.DNI occurred in 26.1% of patients with ALVO who did not experience ENI after EVT. Our study identified several independent predictors of DNI that should be highly considered in daily clinical practice to improve ALVO management.

We aimed to explore a non-invasive estimate of pressure drop in patients who undergo venous sinus stenting to treat idiopathic intracranial hypertension (IIH).This prospective study included 28 IIH patients scheduled for venous stenting. 4D-flow MRI was acquired 24-48 h before venous manometry. Manometry-obtained pressure drop (Mp) was dichotomized into low (Lp: 0-8 mmHg) and high (Hp: 8-30 mmHg) groups. Hemodynamic indices were compared between Lp and Hp. Trans-stenotic pressure drop was estimated by work-energy equation, simplified Bernoulli equation, vorticity magnitude, and velocity difference between inlet and outlet and was compared with Mp. Measurement agreement, correlation, and accuracy were evaluated using the κ coefficient, Pearson's r, and confusion matrix-derived accuracy.Among 28 patients (mean age 38.8 ± 12.7), 19 (67.9%) were female. Work-energy equation-estimated pressure drop (WEp) had strong correlation (r = 0.91, 95% confidence interval [CI]: 0.81-0.96, p0.001) and high agreement (intraclass correlation coefficient = 0.90, 95% CI: 0.78-0.95, p0.001) with Mp. WEp classified Lp and Hp with an accuracy of 0.96. The κ value between WEp and Mp was 0.92 (95% CI: 0.78-1.00). In the work-energy equation, the viscosity energy term (VTrans-stenotic pressure drop in IIH can be estimated using the work-energy equation with favorable accuracy.• Trans-stenotic pressure drop in patients with idiopathic intracranial hypertension can be estimated accurately with the work-energy equation using the 4D-flow MRI full velocity field. • Compared with traditional venous sinus manometry, the 4D-flow MRI-derived pressure drop is totally non-invasive and cost-saving. • 4D-flow MRI may help neurointerventionalist to select IIH patients suitable for venous sinus stenting.

Over the last half century, the main application of Brain Computer Interfaces, BCIs has been controlling wheelchairs and neural prostheses or generating text or commands for people with restricted mobility. There has been very limited attention in the field to applications for computer aided design, despite the potential of BCIs to provide a new form of environmental interaction. In this paper we introduce the development and application of Neuron, a novel BCI tool that enables designers with little experience in neuroscience or computer programming to gain access to neurological data, along with established metrics relevant to design, create BCI interaction prototypes, both with digital onscreen objects and physical devices, and evaluate designs based on neurological information and record measurements for further analysis. After discussing the BCI tool development, the article presents its capabilities through two case studies, along with a brief evaluation of the tool performance and a discussion of implications, limitations, and future improvement.

This paper responds to the significant challenges to Integrated Water Resource Management (IWRM) and clarifies its practical value as a theoretical framework. Through thorough theoretical analysis, this paper examines how IWRM can be applied to distinguish water resource management (WRM) as an observable domain with an open boundary and foster sustainable ecosystem management by emphasizing the functional differentiation of WRM. Compared to the examination made solely from the WRM perspective, an IWRM perspective can better develop the structural complexity of the WRM domain and help expose misunderstandings regarding integration issues. Most importantly, IWRM is not merely an explanatory theoretical framework but also a defensive one. It can facilitate structural coupling between WRM and other domains to reduce conflicts and enable concerted action. However, this role of IWRM has been overlooked for a long time.

This paper responds to the significant challenges to Integrated Water Resource Management (IWRM) and clarifies its practical value as a theoretical framework. Through thorough theoretical analysis, this paper examines how IWRM can be applied to distinguish water resource management (WRM) as an observable domain with an open boundary and foster sustainable ecosystem management by emphasizing the functional differentiation of WRM. Compared to the examination made solely from the WRM perspective, an IWRM perspective can better develop the structural complexity of the WRM domain and help expose misunderstandings regarding integration issues. Most importantly, IWRM is not merely an explanatory theoretical framework but also a defensive one. It can facilitate structural coupling between WRM and other domains to reduce conflicts and enable concerted action. However, this role of IWRM has been overlooked for a long time.

Based on sedimentary theory and high-resolution sequence stratigraphic theory, Denglouku third member in the south of Shuangcheng area is finely divided into a long-term reference surface cycle, 3 medium-term reference surface cycles and 20 short-term reference surface cycles. Denglouku third member is divided into three sand groups from top to bottom according to the medium-term reference surface cycles, and 6 layers, 6 layers, 8 layers in the groups respectively. Through fine observation of rock cores and data analysis of nearly one hundred wells, it is believed that alluvial fan facies, fluvial facies and delta facies are developed in the study area, which can be further divided into 6 subfacies and 12 microfacies. After the analysis to the plane distribution and vertical evolution characteristics of sedimentary facies, it is found that Denglouku third member in the south of Shuangcheng area shows a water-transgression sedimentary cycle from bottom to top which reflects the sedimentary process of water body gradually deepening.

End-to-end speech-to-text translation (E2E-ST) is becoming increasingly popular due to the potential of its less error propagation, lower latency, and fewer parameters. Given the triplet training corpus $\langle speech, transcription, translation\rangle$, the conventional high-quality E2E-ST system leverages the $\langle speech, transcription\rangle$ pair to pre-train the model and then utilizes the $\langle speech, translation\rangle$ pair to optimize it further. However, this process only involves two-tuple data at each stage, and this loose coupling fails to fully exploit the association between triplet data. In this paper, we attempt to model the joint probability of transcription and translation based on the speech input to directly leverage such triplet data. Based on that, we propose a novel regularization method for model training to improve the agreement of dual-path decomposition within triplet data, which should be equal in theory. To achieve this goal, we introduce two Kullback-Leibler divergence regularization terms into the model training objective to reduce the mismatch between output probabilities of dual-path. Then the well-trained model can be naturally transformed as the E2E-ST models by the pre-defined early stop tag. Experiments on the MuST-C benchmark demonstrate that our proposed approach significantly outperforms state-of-the-art E2E-ST baselines on all 8 language pairs, while achieving better performance in the automatic speech recognition task. Our code is open-sourced at https://github.com/duyichao/E2E-ST-TDA., Comment: AAAI 2022

Recently many efforts have been devoted to applying graph neural networks (GNNs) to molecular property prediction which is a fundamental task for computational drug and material discovery. One of major obstacles to hinder the successful prediction of molecular property by GNNs is the scarcity of labeled data. Though graph contrastive learning (GCL) methods have achieved extraordinary performance with insufficient labeled data, most focused on designing data augmentation schemes for general graphs. However, the fundamental property of a molecule could be altered with the augmentation method (like random perturbation) on molecular graphs. Whereas, the critical geometric information of molecules remains rarely explored under the current GNN and GCL architectures. To this end, we propose a novel graph contrastive learning method utilizing the geometry of the molecule across 2D and 3D views, which is named GeomGCL. Specifically, we first devise a dual-view geometric message passing network (GeomMPNN) to adaptively leverage the rich information of both 2D and 3D graphs of a molecule. The incorporation of geometric properties at different levels can greatly facilitate the molecular representation learning. Then a novel geometric graph contrastive scheme is designed to make both geometric views collaboratively supervise each other to improve the generalization ability of GeomMPNN. We evaluate GeomGCL on various downstream property prediction tasks via a finetune process. Experimental results on seven real-life molecular datasets demonstrate the effectiveness of our proposed GeomGCL against state-of-the-art baselines.

Background: The role of stroke etiology subtype in patients with acute large vessel occlusion on the occurrence of hemorrhagic transformation (HT) after endovascular treatment is poorly studied, and which factors mediate their relationship remains largely unknown. We utilized nationwide registry data to explore the association of stroke subtype (cardioembolism versus large artery atherosclerosis) with HT and to identify the possible mediators. Methods: A total of 1015 subjects were selected from the ANGEL-ACT registry (Endovascular Treatment Key Technique and Emergency Work Flow Improvement of Acute Ischemic Stroke)—a prospective consecutive cohort of acute large vessel occlusion patients undergoing endovascular treatment at 111 hospitals in China between November 2017 and March 2019—and divided into large artery atherosclerosis (n=538) and cardioembolism (n=477) according to the Trial of ORG 10172 in Acute Stroke Treatment criteria. The types of HT included any intracerebral hemorrhage (ICH), parenchymal hematoma, and symptomatic ICH within 24 hours after endovascular treatment. The association between stroke subtype and HT was analyzed using a logistic regression model. Mediation analysis was done to assess how much of the effect of stroke subtype on HT was mediated through the identified mediators. Results: Stroke subtype (cardioembolism versus large artery atherosclerosis) was associated with increased risk of any ICH (29.8% versus 16.5%; odds ratio, 2.03 [95% CI, 1.22–3.36]), parenchymal hematoma (14.3% versus 5.4%; odds ratio, 2.90 [95% CI, 1.38–6.13]), and symptomatic ICH (9.9% versus 4.7%; odds ratio, 2.59 [95% CI, 1.09–6.16]) after adjustment for potential confounders. The more thrombectomy passes in cardioembolism patients had a significant mediation effect on the association of stroke subtype with increased risk of HT (any ICH, 15.9%; parenchymal hematoma, 13.4%; symptomatic ICH, 14.2%, respectively). Conclusions: Stroke subtype is an independent risk factor for HT within 24 hours following endovascular treatment among acute large vessel occlusion patients. Mediation analyses propose that stroke subtype contributes to HT partly through thrombectomy pass, suggesting a possible pathomechanistic link. Registration: URL: https://www.clinicaltrials.gov ; Unique identifier: NCT03370939.

With an uptrend of product miniaturization and precision manufacturing, the microembossing process of sheet metal has been widely used, and the corresponding researches have been conducted from the aspects of both experimental and theoretical explorations. Many unknowns, however, remain, such as size effect and it-induced scatters of deformation behaviors and product qualities. In this research, the multi-channel structures with various thicknesses made by a microembossing system using pure copper sheets were employed as case studies. The crystal plasticity finite element method simulations and experiments were conducted to study the effects of grain size and sheet thickness on the process and product of microembossing in terms of load-stroke relation, dimensional accuracy, and surface quality. From both the experimental and numerical results, it is revealed that the grain size effect causes deviated load and non-uniformity in deformation. The dimensional accuracy designated by the variations in thickness and height is affected by both grain size and sheet thickness. As grain size increases, the thickness reduction becomes more dramatic, and the thinnest location is at the corner zone. The height of the channel bottom increases with grain size, while the peak height fluctuation due to grain size is relatively minor. In addition, various surface defects on the surface of the embossed multi-channel structure were observed, including microvoids and micro scratches on the sidewalls, and wrinkles and grooves on the peak and bottom surfaces. From the inner to outer of the structures, microvoids gradually change to micro scratches, which become denser and longer as grain size increases in 100 μm thick copper sheets and also more severe with the increasing sheet thickness. All of these enrich the understanding of sheet metal microembossing and facilitate design solution generation in the optimization of the microembossing process and the quality of the embossed structures.

In today’s contemporary civilization, there is a growing need for clean energy focused on preserving the environment; thus, dielectric capacitors are crucial equipment in energy conversion. On the other hand, the energy storage performance of commercial BOPP (Biaxially Oriented Polypropylene) dielectric capacitors is relatively poor; hence, enhancing their performance has drawn the attention of an increasing number of researchers. This study used heat treatment to boost the performance of the composite made from PMAA and PVDF, combined in various ratios with good compatibility. The impacts of varying percentages of PMMA-doped PMMA/PVDF mixes and heat treatment at varying temperatures were systematically explored for their influence on the attributes of the blends. After some time, the blended composite’s breakdown strength improves from 389 kV/mm to 729.42 kV/mm at a processing temperature of 120 °C. Consequently, the energy storage density is 21.12 J/cm3, and the discharge efficiency is 64.8%. The performance has been significantly enhanced compared to PVDF in its purest state. This work offers a helpful technique for designing polymers that perform well as energy storage materials.

Generating facial reactions in a human-human dyadic interaction is complex and highly dependent on the context since more than one facial reactions can be appropriate for the speaker's behaviour. This has challenged existing machine learning (ML) methods, whose training strategies enforce models to reproduce a specific (not multiple) facial reaction from each input speaker behaviour. This paper proposes the first multiple appropriate facial reaction generation framework that re-formulates the one-to-many mapping facial reaction generation problem as a one-to-one mapping problem. This means that we approach this problem by considering the generation of a distribution of the listener's appropriate facial reactions instead of multiple different appropriate facial reactions, i.e., 'many' appropriate facial reaction labels are summarised as 'one' distribution label during training. Our model consists of a perceptual processor, a cognitive processor, and a motor processor. The motor processor is implemented with a novel Reversible Multi-dimensional Edge Graph Neural Network (REGNN). This allows us to obtain a distribution of appropriate real facial reactions during the training process, enabling the cognitive processor to be trained to predict the appropriate facial reaction distribution. At the inference stage, the REGNN decodes an appropriate facial reaction by using this distribution as input. Experimental results demonstrate that our approach outperforms existing models in generating more appropriate, realistic, and synchronized facial reactions. The improved performance is largely attributed to the proposed appropriate facial reaction distribution learning strategy and the use of a REGNN. The code is available at https://github.com/TongXu-05/REGNN-Multiple-Appropriate-Facial-Reaction-Generation.

With the increasing demand for intelligent services of online video platforms, video character search task has attracted wide attention to support downstream applications like fine-grained retrieval and summarization. However, traditional solutions only focus on visual or coarse-grained social information and thus cannot perform well when facing complex scenes, such as changing camera view or character posture. Along this line, we leverage social information and scene context as prior knowledge to solve the problem of character search in complex scenes. Specifically, we propose a scene-prior-enhanced framework, named SoCoSearch. We first integrate multimodal clues for scene context to estimate the prior probability of social relationships, and then capture characters' co-occurrence to generate an enhanced social context graph. Afterwards, we design a social context-aware GCN framework to achieve feature passing between characters to obtain robust representation for the character search task. Extensive experiments have validated the effectiveness of SoCoSearch in various metrics., Accepted by ICME 2023

ObjectiveSarcopenia is a typical age-related disorder characterized by loss of muscle mass, strength, and physical function. Resistance training has a noticeable effect on sarcopenia, but there is no consensus on whether nutritional supplements can boost this effect. We conducted a meta-analysis of relevant literature to investigate the therapeutic effect of resistance training combined with nutrition intervention on sarcopenia compared with resistance training alone.MethodsCochrane Library, PubMed, Web of Science, Embase, Sinomed, CNKI, VIP, and Wanfang Data were searched for relevant studies on resistance training combined with nutritional intervention for aging adults with sarcopenia. The retrieval period ranged from the inception of the databases to May 24, 2022. Literature screening and information extraction were performed by two researchers. The Physiotherapy Evidence Database (PEDro) scale was adopted for literature quality evaluation and Stata 15.0 software for analysis.ResultsTwelve clinical trials were included, involving 713 older adults diagnosed with sarcopenia, of whom 361 were assigned to the experimental group and 352 to the control group. Compared with the control group, grip strength of the experimental group was substantially elevated [WMD = 1.87, 95% CI (0.01, 3.74), P = 0.049]. Subgroup analysis demonstrated that vitamin D and protein increased grip strength and gait speed. There were no significant improvement in grip strength and gait speed in the protein and vitamin D free subgroup.ConclusionsThis meta-analysis demonstrated that resistance training combined with additional nutritional supplementation, especially compound nutritional supplements that included protein and vitamin D, might further enhance grip strength rather than muscle mass in older adults with sarcopenia.Systematic review registrationhttps://www.crd.york.ac.uk/PROSPERO/, identifier: CRD42022346734.

BackgroundControversy exists regarding the need of advanced imaging for patient selection in the extended window.AimsTo analyze the effect of initial imaging modalities on clinical outcomes of patients underwent MT in the extended window.MethodsThis was a retrospective analysis of a prospective registry, the Endovascular Treatment Key Technique and Emergency Workflow Improvement of Acute Ischemic Stroke (ANGEL-ACT) registry which was conducted at 111 hospitals between November 2017 and March 2019 in China. Primary study cohort and Guideline like cohort were identified, in each cohort, two imaging modalities for patient selection in 6 to 24 h window were defined: (1) NCCT ± CTA, (2) MRI. Guideline-like cohort were further screened based on key features of the DAWN and DEFUSE 3 trials. The primary outcome was 90 day mRS. The safety outcomes were sICH, any ICH and 90-day mortality.ResultsAfter adjusting for covariates, there were no significant differences in 90 day mRS or any safety outcomes between two imaging modalities groups in both cohorts. All outcome measures of mixed-effects logistic regression model were consistent with propensity score matching model.ConclusionOur results indicate that patients presented with anterior large vessel occlusion in the extended time window can potentially benefit from MT even in the absence of MRI selection. This conclusion needs to be verified by the prospective randomized clinical trials.

BackgroundResearch on asthma control levels and quality of life is essential for children with asthma during their growth stage. Therefore, it is necessary to develop a questionnaire that can be used for monitoring and evaluating the disease control effectiveness and quality of life of children with asthma in China and to conduct a preliminary evaluation for its reliability, validity, and discriminative ability.MethodsThe questionnaire was created through a literature review and qualitative interviews for a targeted population. Based on the previous work, 30 caregivers of children with asthma and 5 experienced pediatricians reviewed and discussed a collection of items. Then, 72 items were screened and selected to form the draft questionnaire. After three rounds of investigation (with 240, 503, and 360 participants, respectively), the final questionnaire was established according to the evaluation results. The structure of the questionnaire was explored through confirmatory factor analysis. Exploratory factor analysis and variability analysis were applied based on the first two rounds of investigation. Reliability, construct validity, and discriminative ability were evaluated based on the third round of investigation.ResultsThe questionnaire contains 6 dimensions and 34 items, and the total cumulative variance contribution rate was 54.96%; Cronbach's α coefficient was 0.91; the split-half reliability coefficient was 0.75, and the test–retest reliability coefficient was 0.74. The children's age, gender, residence, asthma attack in the last three months, caregivers' education background, and monthly income per caregiver were correlated with patient-reported outcomes of children with asthma.ConclusionThe questionnaire appeared to have good reliability, construct validity, and discriminative ability in children with asthma in China.

Automatic Micro-Expression (ME) spotting in long videos is a crucial step in ME analysis but also a challenging task due to the short duration and low intensity of MEs. When solving this problem, previous works generally lack in considering the structures of human faces and the correspondence between expressions and relevant facial muscles. To address this issue for better performance of ME spotting, this paper seeks to extract finer spatial features by modeling the relationships between facial Regions of Interest (ROIs). Specifically, we propose a graph convolutional-based network, called Action-Unit-aWare Graph Convolutional Network (AUW-GCN). Furthermore, to inject prior information and to cope with the problem of small datasets, AU-related statistics are encoded into the network. Comprehensive experiments show that our results outperform baseline methods consistently and achieve new SOTA performance in two benchmark datasets,CAS(ME)^2 and SAMM-LV. Our code is available at https://github.com/xjtupanda/AUW-GCN., Accepted by ICME-2023

BackgroundThe effects of angioplasty on intracranial atherosclerotic disease (ICAD)-related acute large-vessel occlusion stroke (LVOS) are unknown. We analyzed the efficacy and safety of angioplasty or stenting for ICAD-related LVOS and the optimal treatment duration.MethodsPatients with ICAD-related LVOS from a prospective cohort of the Endovascular Treatment Key Technique and Emergency Work Flow Improvement of Acute Ischemia Stroke registry were classified as follows: the early intraprocedural angioplasty and/or stenting (EAS) group was defined as the strategy using angioplasty or stenting without mechanical thrombectomy (MT) or one attempt of MT; the non-angioplasty and/or stenting (NAS) group, MT procedure without any angioplasty; and the late intraprocedural angioplasty and/or stenting (LAS) group, using same angioplasty techniques following two or more passes of MT. The primary endpoint was the modified Rankin Scale (mRS) score at 90 days. Other efficacy outcomes included mRS scores 0–1, mRS 0–2, and successful recanalization. Death within 90 days, and symptomatic ICH were safety endpoints. We use propensity score method to diminish the effect of treatment-selection bias. The odds ratio of recanalization rate and mRS score among EAS, NAS, and LAS groups were examined by unadjusted and adjusted logistic regression analysis among unweighted samples and inverse probability of treatment weighting (IPTW) samples.ResultsWe divided 475 cases into three groups. Functional outcomes at 90 days were better in the EAS group than in the NAS and LAS groups. The proportion of mRS 0–1, mRS 0–2, and successful recanalization cases were the highest in the EAS group. However, after IPTW, mortality rate among the three groups were similar (EAS vs. NAS vs. LAS: 19.0 vs. 18.1 vs. 18.7%, p = 0.98) as well as symptomatic intracranial hemorrhage within 24 h however, mortality rate and symptomatic intracranial hemorrhage among the three groups were similar. Logistic regression analysis in unweighted samples and IPTW samples both showed that EAS group had better outcomes. IPTW-adjusted logistic regression analysis demonstrated that the EAS group had better outcomes (mRS 0–1) than the NAS group (adjusted odds ratio [aOR], 0.55; 95% confidence interval [CI]: 0.34–0.88, p = 0.01) and LAS (aOR, 0.39; 95% CI: 0.22–0.68, p = 0.001).ConclusionsAngioplasty and/or stenting should be performed at an early stage for ICAD-related acute LVOS.RegistrationURL: https://www.clinicaltrials.gov; Unique identifier: NCT03370939.

Jinping Neutrino Experiment plans to deploy a new type of 8-inch MCP-PMT with high photon detection efficiency for MeV-scale neutrino measurements. This work studies the performance of the MCP-PMTs, including the photon detection efficiency, the charge resolution of the single photoelectron, the transition time spread, single photoelectron response, rates of dark counts and after pulses. We find a long tail in the charge distribution, and combined with the high photon detection efficiency, the overall energy resolution sees substantial improvements. Those results will be provided as the inputs to detector simulation and design. Our results show that the new PMT satisfies all the requirements of the Jinping Neutrino Experiment.

We design a Blatter radical-based molecular spintronic device, and investigate its spin-polarized transport properties using density functional theory and non-equilibrium Green’s function technique. High-performance spin-rectifying and spin-filtering effects are realized. The physical mechanism is explained by the spin-resolved bias voltage-dependent transmission spectra, the energy levels of corresponding molecular projected self-consistent Hamiltonian orbitals, and their spatial distributions. The results demonstrate that the Blatter radical has great potential in the development of high-performance multifunctional molecular spintronic devices.

This study aimed to evaluate the effect of adjunctive melatonin supplementation on clinical outcomes after non-surgical periodontal treatment.PubMed, Embase, and Web of Science databases were systematically searched for randomised controlled trials (RCTs) of melatonin adjuvant therapy for periodontitis from inception until May 2021. The systematic review was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines and registered on The International Prospective Register of Systematic Reviews (PROSPERO) (CRD42021250630). The risk of bias of included studies was assessed according to the Cochrane Handbook for Systematic Reviews of Interventions. The pooled effect estimates were calculated by a random-effects model, and results were expressed as weighted mean differences (WMD).Seven RCTs comprising 412 participants were included in the meta-analysis. The pooled results showed that adjuvant use of melatonin for non-surgical periodontal treatment significantly improved the probing depth (PD) [WMD = - 1.18, 95% CI (- 1.75, - 0.62) IWe found that adjunctive melatonin supplementation can significantly improve the periodontal status after non-surgical treatment, suggesting that melatonin may be a new adjuvant therapy for periodontitis when non-surgical periodontal treatment alone cannot achieve the desired improvement.

In this study, NiCr2O4 coatings co-doped with metal borides MBx (M = V, Mo and Fe) and MnO2 were synthesized by atmospheric plasma spraying. The phase composition, microstructure, infrared radiation performances as well as the thermal resistance of the coatings were investigated. All coatings possessed typically layer stacked microstructure, and the thickness and roughness (Ra) of the ceramic coatings ranging about 50-60 μm and 3.51-3.65 μm, correspondingly. Besides, the doped coatings still had a spinel structure. Since the metal ions and boron ions entered into the spinel lattice, which induced lattice distortion and enhanced the infrared activity of lattice vibrational absorption, the infrared emissivity of the coatings at 5-25 μm was increased effectively. The highest emissivity at 5-25 μm was obtained by V-doped coating, which was 0.8836 about 0.0175 higher than that of pure NiCr2O4 coating. However, the emissivity at 3∼5 μm was declined owing to compensating effect of high-valence ion doping, which reduced the hole's concentration and the intersub-band transition absorption. The lowest emissivity of 0.8664 was obtained with Mo-doped coating at 3-5 μm, which was 0.017 lower than that of pure NiCr2O4 coating. Attributed to the donor levels introduced through high-valance doping, the free electron's concentration increased, which promoted the transition absorption of free electrons from the impurity levels to the conduction band, resulting in higher emissivity at 0.75-2.5 μm. The highest emissivity of 0.935 at 0.75-2.5 μm was obtained with the Mo-doped coating, which was 0.02 higher than that of pure NiCr2O4 coating. As for high temperature performance, the VB2–MnO2 co-doped coating possessed the highest emissivity of 0.92 in 2.5∼25 μm at 1000 °C, and also showed good thermal resistance after 30 thermal cycles with water-cooling from 750 °C to 25 °C.

Background: Sanghuangporus baumii is a traditional Chinese medicine with anti- cancer, anti-tumor, and anti-inflammatory effects. Triterpenoids are one of the main medicinal ingredients found in S. baumii. However, the dynamic changes of triterpenoids content and its molecular regulation mechanism are still unclear. Objective: Some studies have shown that Lanosterol synthase (LS) is a key enzyme involved in the mevalonate pathway (MVA pathway) to produce lanosterol, which is a precursor for synthesizing S. baumii triterpenoids. Therefore, the study of LS gene and expression characteristics can provide clues for the further study of triterpenoids synthesis. Methods: The PCR, RACE PCR, RT-PCR, homologous recombination and prokaryotic expression technology were used to research the gene characteristic and dynamic changes of LS transcription level. Results: The S. baumii LS sequence included a 5’-untranslated region (129 bp), a 3’-untranslated region (87 bp), and an open reading frame (2,229 bp) encoding 734 amino acids. The S. baumii LS protein was expressed in E. coli BL21 (DE3). The transcription start site of the S. baumii LS promoter sequence ranged from 1 740 bp to 1790 bp. The LS promoter contained 12 CAAT-boxes, 5 ABREs, 6 G-Boxes, 6 CGTCA-motifs, and so on. The LS transcription levels were the highest on day 11 in mycelia (1.6-fold), and the triterpenoids content also gradually increased. The transcription levels began to decrease on day 13, but the triterpenoids content still increased. Results: The S. baumii LS sequence included a 5’-untranslated region (129 bp), a 3’-untranslated region (87 bp), and an open reading frame (2,229 bp) encoding 734 amino acids. The S. baumii LS protein was expressed in E. coli BL21 (DE3). The transcription start site of the S. baumii LS promoter sequence ranged from 1 740 bp to 1790 bp. The LS promoter contained 12 CAAT-boxes, 5 ABREs, 6 G-Boxes, 6 CGTCA-motifs, and so on. The LS transcription levels were the highest on day 11 in mycelia (1.6-fold), and the triterpenoids content also gradually increased. The transcription levels began to decrease on day 13, but the triterpenoids content still increased. Conclusion: The S. baumii LS was cloned and characterized to help to understand the mechanism of triterpenoids synthesis. In addition, we studied the relationship between LS transcription level and triterpenoid dynamic accumulation, and we found that they had a certain correlation.

The probe COZ-DNBS displayed excellent selectivity, a fast response, high sensitivity, a large Stokes shift, and bright-yellow emission in response to H2S.

BackgroundIn patients with acute ischemic stroke, hemorrhagic transformation (HT) is a common complication after intravenous thrombolysis (IVT). In this study, we evaluated the relationship between the ratio of C-reactive protein to albumin (CAR) before thrombolysis, HT, and functional outcomes in patients with acute ischemic stroke.MethodsWe retrospectively analyzed data from 354 patients who received thrombolytic therapy at the Second Affiliated Hospital of the Wenzhou Medical University in China between July 2014 and May 2022. CAR was measured on admission, and HT was identified by cranial computed tomography (CT) within 24–36 h after treatment. Poor outcome was defined as a score on the modified Rankin Scale (mRS) > 2 at discharge. The multivariate logistic regression model was used to investigate the association between CAR, HT, and poor outcome after thrombolysis, respectively.ResultsA total of 354 patients were analyzed, and their median CAR was 0.61 (interquartile range, 0.24–1.28). CAR was significantly higher in the 56 patients (15.8%) who experienced HT than in those who did not (0.94 vs. 0.56, p p p = 0.004). Patients with CAR in the third quartile were more likely to experience poor outcome (OR 3.35, 95% CI 1.32 to 8.51, p = 0.01), as were those in the fourth quartile (OR 7.33, 95% CI 2.62 to 20.50, p ConclusionHigh ratio of C-reactive protein to albumin in individuals with ischemic stroke is associated with an increased risk of HT and poor functional outcomes after thrombolysis.

A good feature representation is the key to image classification. In practice, image classifiers may be applied in scenarios different from what they have been trained on. This so-called domain shift leads to a significant performance drop in image classification. Unsupervised domain adaptation (UDA) reduces the domain shift by transferring the knowledge learned from a labeled source domain to an unlabeled target domain. We perform feature disentanglement for UDA by distilling category-relevant features and excluding category-irrelevant features from the global feature maps. This disentanglement prevents the network from overfitting to category-irrelevant information and makes it focus on information useful for classification. This reduces the difficulty of domain alignment and improves the classification accuracy on the target domain. We propose a coarse-to-fine domain adaptation method called Domain Adaptation via Feature Disentanglement~(DAFD), which has two components: (1)the Category-Relevant Feature Selection (CRFS) module, which disentangles the category-relevant features from the category-irrelevant features, and (2)the Dynamic Local Maximum Mean Discrepancy (DLMMD) module, which achieves fine-grained alignment by reducing the discrepancy within the category-relevant features from different domains. Combined with the CRFS, the DLMMD module can align the category-relevant features properly. We conduct comprehensive experiment on four standard datasets. Our results clearly demonstrate the robustness and effectiveness of our approach in domain adaptive image classification tasks and its competitiveness to the state of the art., 10 pages, 7 figures

Supplemental material, sj-docx-1-msc-10.1177_09691413231174163 for Trends in colorectal cancer screening in the United States, 2012 to 2020 by Yue He, Tong Xu, Jiaxin Fang, Li Tong, Wenhui Gao, Yuan Zhang, Yanfang Wang, Yan Xu, Shunyao Shi, Siyu Liu and Lina Jin in Journal of Medical Screening

Additional file 1.

Supplemental material, sj-pdf-1-ibe-10.1177_1420326X211013080 for Impacts of ambient pressure on the stability of smoke layers and maximum smoke temperature under ceiling in ventilated tunnels by Tong Xu, Fei Tang, Xuhao Xu and Qing He in Indoor and Built Environment

Large language models (LLMs) have demonstrated powerful capabilities in both text understanding and generation. Companies have begun to offer Embedding as a Service (EaaS) based on these LLMs, which can benefit various natural language processing (NLP) tasks for customers. However, previous studies have shown that EaaS is vulnerable to model extraction attacks, which can cause significant losses for the owners of LLMs, as training these models is extremely expensive. To protect the copyright of LLMs for EaaS, we propose an Embedding Watermark method called EmbMarker that implants backdoors on embeddings. Our method selects a group of moderate-frequency words from a general text corpus to form a trigger set, then selects a target embedding as the watermark, and inserts it into the embeddings of texts containing trigger words as the backdoor. The weight of insertion is proportional to the number of trigger words included in the text. This allows the watermark backdoor to be effectively transferred to EaaS-stealer's model for copyright verification while minimizing the adverse impact on the original embeddings' utility. Our extensive experiments on various datasets show that our method can effectively protect the copyright of EaaS models without compromising service quality., Comment: Accepted by ACL 2023

A Computational Tool to Assess the Longevity of Buildings through Flexibility

Graph Neural Networks (GNNs) have been broadly applied in many urban applications upon formulating a city as an urban graph whose nodes are urban objects like regions or points of interest. Recently, a few enhanced GNN architectures have been developed to tackle heterophily graphs where connected nodes are dissimilar. However, urban graphs usually can be observed to possess a unique spatial heterophily property; that is, the dissimilarity of neighbors at different spatial distances can exhibit great diversity. This property has not been explored, while it often exists. To this end, in this paper, we propose a metric, named Spatial Diversity Score, to quantitatively measure the spatial heterophily and show how it can influence the performance of GNNs. Indeed, our experimental investigation clearly shows that existing heterophilic GNNs are still deficient in handling the urban graph with high spatial diversity score. This, in turn, may degrade their effectiveness in urban applications. Along this line, we propose a Spatial Heterophily Aware Graph Neural Network (SHGNN), to tackle the spatial diversity of heterophily of urban graphs. Based on the key observation that spatially close neighbors on the urban graph present a more similar mode of difference to the central node, we first design a rotation-scaling spatial aggregation module, whose core idea is to properly group the spatially close neighbors and separately process each group with less diversity inside. Then, a heterophily-sensitive spatial interaction module is designed to adaptively capture the commonality and diverse dissimilarity in different spatial groups. Extensive experiments on three real-world urban datasets demonstrate the superiority of our SHGNN over several its competitors., Comment: Accepted by KDD 2023

Supplemental material, sj-pdf-1-ibe-10.1177_1420326X211013080 for Impacts of ambient pressure on the stability of smoke layers and maximum smoke temperature under ceiling in ventilated tunnels by Tong Xu, Fei Tang, Xuhao Xu and Qing He in Indoor and Built Environment

$k$NN-MT is a straightforward yet powerful approach for fast domain adaptation, which directly plugs pre-trained neural machine translation (NMT) models with domain-specific token-level $k$-nearest-neighbor ($k$NN) retrieval to achieve domain adaptation without retraining. Despite being conceptually attractive, $k$NN-MT is burdened with massive storage requirements and high computational complexity since it conducts nearest neighbor searches over the entire reference corpus. In this paper, we propose a simple and scalable nearest neighbor machine translation framework to drastically promote the decoding and storage efficiency of $k$NN-based models while maintaining the translation performance. To this end, we dynamically construct an extremely small datastore for each input via sentence-level retrieval to avoid searching the entire datastore in vanilla $k$NN-MT, based on which we further introduce a distance-aware adapter to adaptively incorporate the $k$NN retrieval results into the pre-trained NMT models. Experiments on machine translation in two general settings, static domain adaptation and online learning, demonstrate that our proposed approach not only achieves almost 90% speed as the NMT model without performance degradation, but also significantly reduces the storage requirements of $k$NN-MT., Comment: ICLR 2023

Although remarkable progress on the neural table-to-text methods has been made, the generalization issues hinder the applicability of these models due to the limited source tables. Large-scale pretrained language models sound like a promising solution to tackle such issues. However, how to effectively bridge the gap between the structured table and the text input by fully leveraging table information to fuel the pretrained model is still not well explored. Besides, another challenge of integrating the deliberation mechanism into the text-to-text pretrained model for solving the table-to-text task remains seldom studied. In this paper, to implement the table-to-text generation with pretrained language model, we propose a table structure understanding and text deliberating approach, namely TASD. Specifically, we devise a three-layered multi-head attention network to realize the table-structure-aware text generation model with the help of the pretrained language model. Furthermore, a multi-pass decoder framework is adopted to enhance the capability of polishing generated text for table descriptions. The empirical studies, as well as human evaluation, on two public datasets, validate that our approach can generate faithful and fluent descriptive texts for different types of tables.

Finding multiple temporal relationships among locations can benefit a bunch of urban applications, such as dynamic offline advertising and smart public transport planning. While some efforts have been made on finding static relationships among locations, little attention is focused on studying time-aware location relationships. Indeed, abundant location-based human activities are time-varying and the availability of these data enables a new paradigm for understanding the dynamic relationships in a period among connective locations. To this end, we propose to study a new problem, namely multi-Temporal relationship inference among locations (Trial for short), where the major challenge is how to integrate dynamic and geographical influence under the relationship sparsity constraint. Specifically, we propose a solution to Trial with a graph learning scheme, which includes a spatially evolving graph neural network (SEENet) with two collaborative components: spatially evolving graph convolution module (SEConv) and spatially evolving self-supervised learning strategy (SE-SSL). SEConv performs the intra-time aggregation and inter-time propagation to capture the multifaceted spatially evolving contexts from the view of location message passing. In addition, SE-SSL designs time-aware self-supervised learning tasks in a global-local manner with additional evolving constraint to enhance the location representation learning and further handle the relationship sparsity. Finally, experiments on four real-world datasets demonstrate the superiority of our method over several state-of-the-art approaches., Comment: Accepted by KDD 2023. Code and data: https://github.com/agave233/SEENet

Supplemental material, sj-docx-1-msc-10.1177_09691413231174163 for Trends in colorectal cancer screening in the United States, 2012 to 2020 by Yue He, Tong Xu, Jiaxin Fang, Li Tong, Wenhui Gao, Yuan Zhang, Yanfang Wang, Yan Xu, Shunyao Shi, Siyu Liu and Lina Jin in Journal of Medical Screening

As one of the most important psychic stress reactions, micro-expressions (MEs), are spontaneous and transient facial expressions that can reveal the genuine emotions of human beings. Thus, recognizing MEs (MER) automatically is becoming increasingly crucial in the field of affective computing, and provides essential technical support in lie detection, psychological analysis and other areas. However, the lack of abundant ME data seriously restricts the development of cutting-edge data-driven MER models. Despite the recent efforts of several spontaneous ME datasets to alleviate this problem, it is still a tiny amount of work. To solve the problem of ME data hunger, we construct a dynamic spontaneous ME dataset with the largest current ME data scale, called DFME (Dynamic Facial Micro-expressions), which includes 7,526 well-labeled ME videos induced by 671 participants and annotated by more than 20 annotators throughout three years. Afterwards, we adopt four classical spatiotemporal feature learning models on DFME to perform MER experiments to objectively verify the validity of DFME dataset. In addition, we explore different solutions to the class imbalance and key-frame sequence sampling problems in dynamic MER respectively on DFME, so as to provide a valuable reference for future research. The comprehensive experimental results show that our DFME dataset can facilitate the research of automatic MER, and provide a new benchmark for MER. DFME will be published via https://mea-lab-421.github.io., Comment: 16 pages, 7 figures

This paper presents a novel approach to assessing human lighting adjustment behavior and preference in diverse lighting conditions through the evaluation of emotional feedback and behavioral data using VR. Participants (n= 27) were exposed to different lighting (n=17) conditions with different levels of illuminance and correlated color temperature (CCT) with a randomized order in a virtual office environment. Results from this study significantly advanced our understanding of preferred lighting conditions in virtual reality environments, influenced by a variety of factors such as illuminance, color temperature, order of presentation, and participant demographics. Through a comprehensive analysis of user adjustment profiles, we obtained insightful data that can guide the optimization of lighting design across various settings.