Search

Visual impairment community, especially blind people have a thirst for assistance from advanced technologies for understanding and answering the image. Through the development and intersection between vision and language, Visual Question Answering (VQA) is to predict an answer from a textual question on an image. It is essential and ideal to help blind people with capturing the image and answering their questions automatically. Traditional approaches often utilize the strength of convolution and recurrent networks, which requires a great effort for learning and optimizing. A key challenge in VQA is finding an effective way to extract and combine textual and visual features. To take advantage of previous knowledge in different domains, we propose BERT-RG, the delicate integration of pre-trained models into feature extractors, which relies on the interaction between residual and global features in the image and linguistic features in the question. Moreover, our architecture integrates a stacked attention mechanism that exploits the relationship between textual and visual objects. Specifically, the partial regions of images interact with partial keywords in question to enhance the text-vision representation. Besides, we also propose a novel perspective by considering a specific question type in VQA. Our proposal is significantly meaningful enough to develop a specialized system instead of putting forth the effort to dig for unlimited and unrealistic approaches. Experiments on VizWiz-VQA, a practical benchmark dataset, show that our proposed model outperforms existing models on the VizWiz VQA dataset in the Yes/No question type.

Developing low-cost, efficient, and stable non-precious-metal electrocatalysts with controlled crystal structure, morphology and compositions are highly desirable for hydrogen and oxygen evolution reactions. Herein, a series of phosphorus-doped Fe7S8 nanowires integrated within carbon (P-Fe7S8@C) are rationally synthesized via a one-step phosphorization of one-dimensional (1D) Fe-based organic-inorganic nanowires. The as-obtained P-Fe7S8@C catalysts with modified electronic configurations present typical porous structure, providing plentiful active sites for rapid reaction kinetics. Density functional calculations demonstrate that the doping Fe7S8 with P can effectively enhance the electron density of Fe7S8 around the Fermi level and weaken the Fe-H bonding, leading to the decrease of adsorption free energy barrier on active sites. As a result, the optimal catalyst of P-Fe7S8-600@C exhibits a relatively low overpotential of 136 mV for hydrogen evolution reaction (HER) to reach the current density of 10 mA/cm2, and a significantly low overpotential of 210 mV for oxygen evolution reaction (OER) at 20 mA/cm2 in alkaline media. The work presented here may pave the way to design and synthesis of other prominent Fe-based catalysts for water splitting via electronic regulation.

This review focuses on the prochlorosin and cyanobactin biosynthetic pathways, unusual examples of pathways in cyanobacteria that appear to be diversity generating. In the case of prochlorosins, a single enzyme catalyzes post-translational modifications on an impressive number (10–80) of peptide substrates, resulting in polycyclic peptides of very different sizes and sequences. Like the prochlorosins, the cyanobactins are an extraordinarily diverse group of (mostly) cyclic peptides, but they are decorated further in parallel fashion by substrate-tolerant enzymes. This review discusses the current understanding of catalysis and possible evolutionary models that might explain the perplexing enzymatic tolerance of these systems.

The purpose of this research data is to mention a correlation matrix of data-analytic estimates between the HEXACO personality traits and number of job interviews and job offers through job search networking behavior. The data-analytic correlations are based on 773 graduated students in universities in the North of Vietnam. This data used self-report surveys to collect the data. The data was analyzed using the AMOS software version 22. The obtained data indicated that the level of networking behavior positively affects both number of job interviews and job offers of individuals. Moreover, the acquired data revealed that the personality traits of honesty-humility, extraversion, agreeableness, conscientiousness, and openness to experience are positively related to the level of networking behavior, while emotionality is negatively associated with networking intensity.

We parametrize the set of irreducible characters of the Sylow p-subgroups of the Chevalley groups D 6 ( q ) and E 6 ( q ) , for an arbitrary power q of any prime p. In particular, we establish that the parametrization is uniform for p ≥ 3 in type D 6 and for p ≥ 5 in type E 6 , while the prime 2 in type D 6 and the primes 2, 3 in type E 6 yield character degrees of the form q m / p i which force a departure from the generic situations. Also for the first time in our analysis we see a family of irreducible characters of a classical group of degree q m / p i where i > 1 which occurs in type D 6 .

Doping technology has been proved to be a productive strategy to realize the multi-functionality of materials. Herein, Ni doped SnO2 quantum dots (QDs) are fabricated through a convenient hydrothermal process. In the fabrication process, Vitamin C (VC) acts as a stabilizer to avoid the precipitating of Sn4+(Ni2+), and Na2CO3 serves as a precipitator to react with Sn4+(Ni2+) to produce Ni doped SnO2 QDs. The influence of the Ni doping amount on the photodegradation performance of SnO2 QDs is studied. It was found that the Ni doped SnO2 QDs with Ni/Sn atomic ratio of 1% display the best photocatalytic performace due to the following advantages: (1) the Ni doping can boost light-harvesting and narrow the band gaps of the SnO2 QDs; (2) the SnO2 QDs with ultra small particle size and large specific surface area can supply more reactive centers for photocatalytic reactions. The two important factors contribute to enhancing the photocatalytic performance of the 1% Ni doped SnO2 QDs. As a result, the 1% Ni doped SnO2 QDs photodegrade 91.5% Rhodamine B after being illuminated by solar light for 28 min, obviously surpassing that of their pure counterparts.

One-dimensional (1D) heterostructured photocatalysts with controllable texture properties and compositions have attracted increasing interest owing to their unique optical, structural, and electronic advantages. Herein, 1D Co3O4-SnO2 heteronanorods were rationally designed and synthesized through a facile solution-based approach. Benefiting from both of their heterostructural and compositional characteristics, the resulting Co3O4-SnO2 heteronanorods exhibit high photocatalytic performance for the degradation of Rhodamine B (RhB) under visible-light irridation. In particular, the photocatalyst with a Co3O4/SnO2 mass ratio of 1:1 provides the best photocatalytic performance, which can degrade 90% RhB within 120 min. Besides, several reaction parameters affecting RhB degradation, such as churning time, calcination temperature and pH value, are investigated in detail. The enhanced photocatalytic activity can be attributed to the broadening of absorption spectrum to visible-light regions and the efficient charge separation of photogenerated electron-hole pairs due to the formed p-n heterojunctions. The strategy reported here can be able to expand to fabricate other heterostructured photocatalysts for practical applications in the fields of photocatalysis, water splitting, and solar cells.

Construction of heterostructures with proper band alignment and effective transport and separation of photogenerated charges is highly expected for photocatalysis. In this work, Ni-doped SnO2-SnS2 heterostructures (NiSnSO) are simply prepared by thermal oxidation of Ni-doped hierarchical SnS2 microspheres in the air. When applied for the photodegradation of organic contaminants, these NiSnSO exhibit excellent catalytic performance and stability due to the following advantages: (1) Ni doping leads to the enhancement of light harvesting of SnS2 in the visible light regions; (2) the formed heterojunctions promote the transport and separation of photogenerated electrons from SnS2 to SnO2; (3) Ni-SnO2 quantum dots facilitate the enrichment of reactants, provide more reactive centers and accelerate product diffusion in the reactive centers; (4) the SnS2 hierarchical microspheres constituted by nanoplates provide abundant active sites, high structural void porosity and accessible inner surface to faciliate the catalytic reactions. As a result, the optimized NiSnSO can photodegrade 92.7% methyl orange within 80 min under the irradiation of simulated sunlight, greatly higher than those of pure SnS2 (29.8%) and Ni-doped SnS2 (52.1%). These results reveal that the combination of heteroatom doping and heterostructure fabrication is a very promising strategy to deliver nanomaterials for effectively photocatalytic applications.

Three-dimensional (3D) hierarchical architectures are currently attracting extensive interests owing to their fascinating morphology-dependent properties and potential applications. In this work, we have developed a facile template-free hydrothermal method to synthesize various 3D SnO2 hierarchical superstructures. It is found that the amount of urea, pH value and the pre-oxidation process play a crucial role in the micro/nanostructural evolutions of building blocks and final products. Oriented attachment and self-assembling dual-controlled mechanisms are then proposed to illustrate the formation of 3D hierarchical superstructures. The application in photocatalysis for the degradation of methylene blue (MB) demonstrates that the as-synthesized SnO2 hierarchical superstructures have highly efficient photocatalytic activity for the degradation of organic pollutants. Particularly, the SnO2 hierarchical superstructure constructed by ultrathin nanosheets can degrade 93% of MB within 150 min, which is much higher than that of commercial SnO2 powders. The enhanced photocatalytic performances can be attributed to the high surface-to-volume ratio, the abundant active sites and the efficient separation of photogenerated electrons and holes induced by the unique 3D loose and microporous superstructures. These new insights obtained in this study will be beneficial for the practical applications of oxide semiconductor materials for photocatalytic degradation of organic pollutants.

Background Multiple studies investigating the use of focused cardiac ultrasound (FCU) in lower and middle-income countries and in medically underserved areas of the United States have demonstrated utility in echocardiographic screening algorithms performed by a variety of operators at different levels of training. No study to date has employed previously untrained nurses in a medically underserved setting to identify older adults with cardiac disorders. The aim of this study was to assess the accuracy of nurse-performed FCU to screen adult subjects at a village health center in Vietnam. Methods Vietnamese nurses (N = 8) underwent structured training conducted by sonographers and physicians during an outreach event sponsored by the American Society of Echocardiography Education and Research Foundation. The nurses were trained to detect abnormalities from a single echocardiographic view (parasternal long-axis) with a laptop-sized device and underwent pre- and posttraining testing. Following training, cardiac ultrasound examinations were performed on subjects >50 years of age at a village health center. First, the nurses performed focused cardiac ultrasound using two-dimensional and color Doppler imaging in the parasternal long-axis view using the M7 device and recorded their assessments. Two-dimensional color and spectral Doppler echocardiography was thereafter performed using the same machine by a sonographer (n = 5) or a Vietnamese echocardiography-trained cardiologist (n = 1). Interviews and electrocardiography were performed at the time of FCU. Results Each nurse improved from pre- to posttraining (average improvement in correct answers, 21%; range, 2%–31%). During the scanning phase, nurses' sensitivity, specificity, and accuracy for identifying subjects with any abnormality were 51.5% (85 of 165), 78.1% (82 of 105) and 61.9%, respectively. There were 60 subjects with significant findings (22.2%); all of these subjects had significant abnormalities visible on parasternal long-axis images. Overall sensitivity, specificity, and accuracy for identifying subjects with major abnormalities were 83.3% (50 of 60), 78.1% (164 of 210), and 78.6%, respectively. Nurse-performed FCU demonstrated much higher sensitivity with lower specificity than electrocardiography alone. The combination of nurse-performed FCU plus ECG identified all of the significant findings on echocardiography and increased accuracy to 91.5%. Conclusions Nurses with no prior echocardiographic experience and with limited training can identify patients with significant cardiac abnormalities using FCU with acceptable accuracy. Screening strategies involving FCU may play a role in improving access to health care and triage in underserved areas.

This paper describes how Rule Input Network Generator (RING), a network generation computational tool, can be adopted to generate a variety of complex biochemical reaction networks. The reaction language incorporated in RING allows representation of chemical compounds in biological systems with various structural complexity. Complex molecules such as oligosaccharides in glycosylation pathways can be described using a simplified representation of their monosaccharide building blocks and glycosidic bonds. The automated generation and topological network analysis features in RING also allow for: (1) constructing biochemical reaction networks in a rule-based manner, (2) generating graphical representations of the networks, (3) querying molecules containing a particular structural pattern, (4) finding the shortest synthetic pathways to a user-specified species, and (5) performing enzyme knockout to study their effect on the reaction network. Case studies involving three biochemical reaction systems: (1) Synthesis of 2-ketoglutarate from xylose in bacterial cells, (2) N-glycosylation in mammalian cells, and (3) O-glycosylation in mammalian cells are presented to demonstrate the capabilities of RING for robust and exhaustive network generation and the advantages of its post-processing features.

A heat shield called the thermal protection system (TPS) is an important structure in hypersonic vehicles as it prevents hot air from entering vehicles and potential impacts from space debris. With the increase in demand for low-cost reusable launch vehicles as well as for searching and exploration of new planets in both unmanned and manned missions, the need for developing an effective TPS has increased across many countries. The structural design of TPSs has become more prominent in the early stage of hypersonic vehicle development. Sandwich structures that have the advantages of low density and high performance are integrated into the structural design of an effective TPS. This paper provides a comprehensive review of recent research efforts on sandwich structures for TPSs. The topics discussed in this paper include aspects of structural and material design, mechanical and thermomechanical performances, and manufacturing methods. In particular, we review and discuss the structural design as well as the material design of sandwich structures for different TPS types with various configurations, including corrugated cores, lattice cores, multilayer cores, foams, honeycomb cores, bio-inspired cores. The materials used for the sandwich structures, such as various types of laminated composite, ceramic matrix composite, and metals, are included. We also discuss the performance of the TPS sandwich structures in terms of temperature gradients, deformation limits, and mechanical strengths and provide a discussion on the manufacturing methods of TPS sandwich structures for hypersonic vehicles. Finally, further research directions and challenges of sandwich structures for TPSs are presented.

We prove a posteriori upper and lower bounds for the error estimates when solving the Laplace–Beltrami equation on the unit sphere by using the Galerkin method with spherical splines. Adaptive mesh refinements based on these a posteriori error estimates are used to reduce complexity and computational cost of the corresponding discrete problems. The theoretical results are corroborated by numerical experiments.

In this study, the fracture properties of a piezoelectric stack actuator made of a single crystal (PMN-29PT) were investigated using a three point bending test along with the digital image correlation (DIC) technique. To simplify the model of the piezoelectric stack and minimize the cost, two layers of PMN-29PT and one thin copper electrode were glued together using silver conductive adhesive instead of a real piezoelectric stack with 40 PMN-29PT layers. The PMN-29PT layers were then strongly attached with a two square section acrylic beam forming a beam with the same geometry as the piezoelectric stack. The three point bending test results showed that the force at the crack onset was around 85% of the fracture force. The maximum stress and fracture energy from force-displacement data were measured to be 1.41 MPa and 3.83 J/m 2 , respectively. Furthermore, the fracture surface indicated that the fracture energy was affected by the surface of the copper electrode, and that the rough surface of the electrode can increase the fracture energy. The fracture energy was determined using a simple beam theory. The fracture energy according to the simple beam theory was slightly smaller than that from force-displacement data. Understanding the measured fracture properties of the piezoelectric stack will help us prevent the fracture of the stack when the piezoelectric stack actuator is handled.

Let $q$ be a power of a prime $p$, let $G$ be a finite Chevalley group over $\mathbb{F}_q$ and let $U$ be a Sylow $p$-subgroup of $G$; we assume that $p$ is not a very bad prime for $G$. We explain a procedure of reduction of irreducible complex characters of $U$, which leads to an algorithm whose goal is to obtain a parametrization of the irreducible characters of $U$ along with a means to construct these characters as induced characters. A focus in this paper is determining the parametrization when $G$ is of type $\mathrm{F}_4$, where we observe that the parametrization is "uniform" over good primes $p > 3$, but differs for the bad prime $p = 3$. We also explain how it has been applied for all groups of rank $4$ or less., 38 pages

Cultured mammalian cells are the main workhorses for producing biologics. The performance of these cell culture processes, in terms of both productivity and product quality attributes, is significantly influenced by cellular metabolism. Glucose is the major carbon source for cellular biosynthesis and energy generation. We summarize here recent advances in our understanding of the regulation of glucose metabolism in cultured cells. The versatility of cells to sustain homeostatic states under widely varying environments is made possible by allosteric regulation of the metabolic network, interplay between the signaling pathways, and transcription factors. Understanding the regulation of metabolism holds the key to altering the metabolic regulatory circuit and implementing direct metabolic control over cell culture processes.

We propose in this paper a decentralized traffic signal control policy for urban road networks. Our policy is an adaptation of a so-called BackPressure scheme which has been widely recognized in data network as an optimal throughput control policy. We have formally proved that our proposed BackPressure scheme, with fixed cycle time and cyclic phases, stabilizes the network for any feasible traffic demands. Simulation has been conducted to compare our BackPressure policy against other existing distributed control policies in various traffic and network scenarios. Numerical results suggest that the proposed policy can surpass other policies both in terms of network throughput and congestion., Public technical report

This paper presents the use of the Digital Image Correlation (DIC) technique to investigate the thermal buckling of a circular laminated composite plate subjected to a uniform distribution of temperature load. The circular laminated composite plate made of glass/epoxy was placed in a titanium ring without being fixed and the structure was heated from room temperature 30 °C to 120 °C in the heating chamber. The thermal deformation and strain of the composite plate resulting from the thermal load were measured using DIC. The buckling temperature and the first buckling mode shape were determined from the full-field deformation shape and temperature–displacement curve that were obtained from the experiment. In addition, the buckling temperature of the circular composite plate was also studied using energy theory. The results of the buckling temperature from DIC were close to the theoretical buckling temperature of the circular plate found using a simply supported boundary condition. In order to verify the proposed measurement method, the finite element analysis method was used to perform a nonlinear buckling analysis with the influence of imperfections. These results show that the present method is promising for studying thermal buckling of composite structures in diverse fields such as aerospace engineering.

Global survey of transcriptome dynamics can provide molecular insights into cell physiology. In the past few years, DNA microarray for transcriptome analysis has been augmented by high-throughput sequencing methods; extending the reach of transcriptome analysis to the rodent species of biotechnological importance, for which the development of genomic tools has been lagging. The rapid accumulation of sequencing data for these species highlighted the need for more evidence-based annotation. Recent findings in the epigenetic regulation in human and mouse will inspire similar research in CHO and BHK cells. Transcriptome studies in these recombinant cells will likely lay the foundation for a systems-based genome engineering which can be used to develop superior producing cell lines. Herein, we summarized the recent findings and advances in transcriptome studies of cell culture bioprocesses. The potential impact of transcriptomics on biopharmaceutical process technology is also discussed.

Advancements in the efficiency, quality and manufacturability of sensing and communication systems are driving the field of intelligent transport systems (ITS) into the twenty first century. One key aspect of ITS is the need for efficient and robust integrated network management of urban traffic networks. This paper presents a general model predictive control framework for both centralized traffic signal and route guidance systems aiming to minimize network congestion. Our novel model explicitly captures both non-zero travel time and spill-back constraints while remaining linear and thus generally tractable with quadratic costs. The end result is a central control scheme that may be realized for large urban networks containing thousands of sensors and actuators.We demonstrate the essences of our model and controller through a detailed mathematical description coupled with simulation results of specific scenarios. We show that using a central scheme such as ours may reduce the congestion inside the network by up to half while still achieving better throughput compared to that of other conventional control schemes.

Let U ( q ) be a Sylow p-subgroup of the Chevalley groups D 4 ( q ) where q is a power of a prime p. We describe a construction of all complex irreducible characters of U ( q ) and obtain a classification of these irreducible characters via the root subgroups which are contained in the center of these characters. Furthermore, we show that the multiplicities of the degrees of these irreducible characters are given by polynomials in ( q − 1 ) with nonnegative integer coefficients.

Aqueous phosphoric acid (85 wt%) is an efficient and mild reagent for the deprotection of N-BOC groups. Acid sensitive functionalities including benzyl and methyl esters, TBDMS ether, CBZ and isopropylidene groups are compatible with the reaction conditions. The reactions are high yielding, and the workup is convenient.

The credit channel of monetary transmission suggests that monetary tightening results in a contraction of the supply of credit to firms who borrow via financial intermediaries. However, according to Meltzer (1960), the existence of an inter-firm credit flow appears to favour those firms most affected by credit rationing. This study re-analyzes Meltzer's hypothesis using UK data. The results of the analysis shows that indeed there is an increase in trade credit flows in period of rising interest rates, but the magnitude is not very significant overall. Firms of different financial status, however, behave differently. The novel finding is that small firms extend trade credit more aggressively than medium and large firms. This behaviour occurs even in those firms in financial distress. Interestingly, this behaviour may be one of the reasons that cause small firms that cannot receive trade credit from their suppliers in period of monetary contraction, to go bankrupt.