Refinement Method Research Articles

Theoretical and experimental investigation of BaY2(MoO4)4:xSm3+ phosphors

In this paper, the novel BaY2(MoO4)4:xSm3+ phosphors with strong red emission and high thermal stabilization were prepared by a conventional method of high temperature solid state reaction in air atmosphere. Morphology, phase purity, element distribution, UV–Vis–NIR diffuse reflectance spectroscopy (DRS), and luminescence properties were investigated in detail. The crystal structure of the BaY2(MoO4)4 host was refined with the aid of the Rietveld refinement method by the GSAS program. The electronic band structure and phonon dispersion were performed using plane-wave density functional theory (DFT). The prepared BaY2(MoO4)4:Sm3+ phosphor exhibits excellent red emission under near-ultraviolet excitation. Strong red emission intensity and high thermal stability suggest potential application in backlighting or phosphor-covered white light-emitting diodes (pc-LEDs).

Effective strategies for improving quenched microstructure and mechanical properties of hot-rolled seamless steel pipes

The containing Ti-Mg-Ca-O composite particles were innovatively used to achieve microstructure refinement and properties improvement of hot-rolled seamless steel pipes, and compared with conventional steel (C# steel). The results show that the C# steel mainly consisted of parallel lath martensite after on-line quenching. Under the same rolling and heat treatment conditions, dominant Ti-Mg-Ca-O composite particles in modified steel (M# steel) were effective nucleation sites of acicular ferrite, dividing the prior austenite grains and refining microstructures. The strength values of the both steels were at the same level, however, the impact energy of M# steel was significantly higher than C# steel, increasing by 28%. This study provides an effective method for further microstructure refinement of hot-rolled seamless steel tubes.

GEEES: inferring cell-specific gene-enhancer interactions from multi-modal single-cell data.

Gene-enhancer interactions are central to transcriptional regulation. Current multi-modal single-cell datasets that profile transcriptome and chromatin accessibility simultaneously in a single cell are yielding opportunities to infer gene-enhancer associations in a cell type specific manner. Computational efforts for such multi-modal single-cell datasets thus far focused on methods for identification and refinement of cell types and trajectory construction. While initial attempts for inferring gene-enhancer interactions have emerged, these have not been evaluated against benchmark datasets that materialized from bulk genomic experiments. Furthermore, existing approaches are limited to inferring gene-enhancer associations at the level of grouped cells as opposed to individual cells, thereby ignoring regulatory heterogeneity among the cells. We present a new approach, GEEES for "Gene EnhancEr IntEractions from Multi-modal Single Cell Data," for inferring gene-enhancer associations at the single-cell level using multi-modal single-cell transcriptome and chromatin accessibility data. We evaluated GEEES alongside several multivariate regression-based alternatives we devised and state-of-the-art methods using a large number of benchmark datasets, providing a comprehensive assessment of current approaches. This analysis revealed significant discrepancies between gold-standard interactions and gene-enhancer associations derived from multi-modal single-cell data. Notably, incorporating gene-enhancer distance into the analysis markedly improved performance across all methods, positioning GEEES as a leading approach in this domain. While the overall improvement in performance metrics by GEEES is modest, it provides enhanced cell representation learning which can be leveraged for more effective downstream analysis. Furthermore, our review of existing experimentally driven benchmark datasets uncovers their limited concordance, underscoring the necessity for new high-throughput experiments to validate gene-enhancer interactions inferred from single-cell data. https://github.com/keleslab/GEEES.

Effect of Synthesis Process on The Piezoelectric Properties of (1-x) NBT-xBT: A Comparative Study Between Solid State and Hydrothermal Methods

Lead-free oxide perovskite materials are produced to ensure that no toxic or dangerous waste is released into the environment, although this technique often requires a high-temperature synthesis method. Nevertheless, the development of an environmentally friendly chemical process to carry out the synthesis of perovskite material in the laboratory at low temperature is a major challenge. In this investigation, to lower the temperature of the thermal treatment, (1-x)(Na0.5Bi0.5)TiO3-xBaTiO3 (abbreviated as (1-x)NBT-xBT) ceramics were synthesized using two different methods: a hydrothermal method and a conventional calcination process (solid-state), using the same reagents. The effects of various synthesis techniques on the composition, structure, morphology, and dielectric properties of the (1-x)NBT-xBT ceramics were registered and examined. The XRD plots taken at ambient temperature were used to verify the phase formation of the samples. The occurrence of functional bands was also verified by Raman spectroscopy at ambient temperature. Using the Rietveld refinement method were able to detect the morphotropic phase boundary (MPB) near x=0.05-0.07. When the hydrothermal process was adopted, the (1-x)NBT-xBT ceramics show a single perovskite structure sintered at 1000°C. Moreover, scanning Electron Microscopy (SEM) examination revealed a uniform distribution of grains and a significant change in grain size with the increase in BaTiO3 concentration when compared to the conventional method. The microstructure of ceramics is generally strongly influenced by the method used to prepare the powders, which has a considerable impact on their performance. Piezoelectric and dielectric properties of the specimens have been compared with the properties of those prepared by the hydrothermal process. Thus, the choice of the suitable synthesis method depends on the desired properties of the (1-x)NBT-xBT materials.

Statistical refinement of patient-centered case vignettes for digital health research.

Digital health research often relies on case vignettes (descriptions of fictitious or real patients) to navigate ethical and practical challenges. Despite their utility, the quality and lack of standardization of these vignettes has often been criticized, especially in studies on symptom-assessment applications (SAAs) and self-triage decision-making. To address this, our paper introduces a method to refine an existing set of vignettes, drawing on principles from classical test theory. First, we removed any vignette with an item difficulty of zero and an item-total correlation below zero. Second, we stratified the remaining vignettes to reflect the natural base rates of symptoms that SAAs are typically approached with, selecting those vignettes with the highest item-total correlation in each quota. Although this two-step procedure reduced the size of the original vignette set by 40%, comparing self-triage performance on the reduced and the original vignette sets, we found a strong correlation (r = 0.747 to r = 0.997, p < .001). This indicates that using our refinement method helps identifying vignettes with high predictive power of an agent's self-triage performance while simultaneously increasing cost-efficiency of vignette-based evaluation studies. This might ultimately lead to higher research quality and more reliable results.

Pixel-wise matching cost function for robust light field depth estimation

Depth estimation has significant potential in industrial applications, such as robot navigation. Light Field (LF) technology captures both spatial and angular information of a scene, enabling precise depth acquisition. Stereo matching technology has been commonly used for this task in LF imaging. However, previous methods typically compute the matching cost by simply shifting images with predefined offsets, causing poor performance in regions with sudden depth distribution changes (e.g., edges). To address this issue, we propose a Pixel-wise Matching Cost Function (PMCF) to estimate depth in units of pixel, then design a pixel-based network, termed PixelNet, for depth prediction, which achieves top rankings on the HCI 4D light field benchmark. Specifically, our cost function consists of two modules: the range search strategy and the modulation mechanism. The range search strategy enables the network to iteratively optimize depth on a per-pixel basis, while the modulation mechanism effectively handles scene noise and occlusions. By integrating these two modules, our function enables the generation of depth maps with sharper edges and smoother surfaces, compared with previous methods. Finally, we design a pixel-based network and apply it to both synthetic and real-world scenes, demonstrating that our method outperforms state-of-the-art methods in both scenes. Furthermore, PixelNet can also function as a post-processing network that can be seamlessly integrated with existing methods for depth refinement.

Structural and Chemical Analysis of Three Regions of Bamboo (Phyllostachys Edulis)

This study focuses on three different regions of moso bamboo (Phyllostachys edulis): an inner layer (IB), middle layer (MB), and outer layer (OB), to comprehensively characterize the structural features, chemical composition (ash, extractives and lignin contents), and the lignin monomeric composition as determined by analytical pyrolysis. The results show that bamboo presents a gradient structure. From the IB to OB, the vascular bundle density and fiber sheath ratio increase, the porosity decreases (from 45.92% to 18.14%), and the vascular bundle diameter–chord ratio increases (from 0.85 to 1.48). In terms of chemical composition, the ash, extractives, and acid-soluble lignin content gradually decrease from IB to OB. The holocellulose content follows the trend: MB (66.3%) &gt; OB (65.9%) &gt; IB (62.8%), while the acid-insoluble lignin content exhibits the opposite trend: IB (22.6%) &gt; OB (17.8%) &gt; MB (17.7%). Pyrolysis products reveal the diversity of carbohydrates and lignin derivatives, with a lignin monomeric composition rich in syringyl and guaiacyl units and lower amounts of H-units: the IB has an H:G:S relation of 18:26:55, while 15:27:58 is the ratio for the MB and 15:40:45 for the OB; S/G ratio values were, respectively, 1.22, 1.46, and 0.99. A comprehensive analysis highlights significant gradient variations in the structure and chemistry of bamboo, providing robust support for the classification and refinement methods of bamboo residues for potential applications.

Learning a CNN with the finite volume method for modelling fluid seepage flow under closed boundary

Physics-informed neural networks (PINNs) are an efficient technique for incorporating physics law in models based on neural networks (NNs). In this paper, we propose a local refinement physics-informed convolutional neural network (LR-PICNN) to solve the seepage equation in closed boundary reservoirs. LR-PICNN employs a local refinement method to enhance focus on the region containing the well. First, the finite volume (FV) method is used to discretize the seepage equation on the global reservoir domain and on the local domain around the well to construct the loss functions respectively. Second, the boundary between the global and local domains are used as boundary conditions to couple the two systems. Finally, two loss functions are added to train the CNN to solve the seepage equation in autoregressive way and coupled without any labeled data. Experimental results show that LR-PICNN can more accurately solve seepage flow equation. Due to the coupling of solving equations and that LR-PICNN focuses on the behaviors of the well, compared to physics-informed Deep Convolutional neural network (PIDCNN) under the same well grid size, LR-PICNN has a faster solving speed that is at least 8 times faster, higher solution accuracy for BHP that is at least twice as high, and more stable solution, which could be used in numerical well testing.

Tin-doped titanium dioxide film-enhanced electrocatalytic hydrogen evolution

Electrocatalyst water-splitting based on metal oxide nanocomposites has gained considerable interest in hydrogen evolution applications. Here, undoped titanium dioxide and tin-doped titanium dioxide films (TiO2 and Sn/TiO2, respectively) are presented for highly efficient H2 production applications. The crystal structure analysis is performed by analyzing the XRD patterns using the Rietveld refinement method and the Williamson-Hall method. XRF scans are used to confirm the doping mechanism between Sn and TiO2. The bandgap energies of undoped TiO2 and Sn/TiO2 films are 3.33 and 3.15 eV, respectively. On the other hand, the electrical conductivity values of undoped TiO2 and Sn/TiO2 films are 0.10 and 0.25 mS.cm−1, respectively. The electrochemical H2 production performance of undoped TiO2 and Sn/TiO2 films is investigated through two different methods: potentiostat measurements and analytical methods. It can be concluded that the Sn/TiO2 film exhibits higher HER performance and H2 production efficiency than the undoped TiO2 film.

Refinement method of equivalent source power level determination for traffic noise in urban noise mapping

For traffic noise prediction, the predictive models of source power level determination commonly in use are the Technical Guidelines for Noise Impact Assessment in China or the Calculation of Road Traffic Noise model in the United Kingdom, which incorporate traffic flow and car speeds. However, the accuracy of these models is often compromised due to urban traffic's complexity, vehicle diversity, and unforeseen events. This study investigates a refinement method based on noise on-site measurement and aforementioned approaches for equivalent source power level determination for traffic noise in urban noise mapping. In the study, onsite measurements are conducted to collect noise and traffic data, and source power level calculated by refinement method is validated using measurement data. Different prediction methods are employed for measurement sites calculations. By comparisons between predictions with noise on-site measurement data, the refinement method proposed in the study demonstrates a significant accuracy enhancement in traffic noise prediction over previous models for determining equivalent source power level. This enhancement highlights the efficacy of adopting noise measurement-based methods for setting traffic noise source power levels, leading to more precise traffic noise predictions in complex urban settings, particularly in areas surrounding urban arterial roads.

Extracting Sentence Embeddings from Pretrained Transformer Models

Pre-trained transformer models shine in many natural language processing tasks and therefore are expected to bear the representation of the input sentence or text meaning. These sentence-level embeddings are also important in retrieval-augmented generation. But do commonly used plain averaging or prompt templates sufficiently capture and represent the underlying meaning? After providing a comprehensive review of existing sentence embedding extraction and refinement methods, we thoroughly test different combinations and our original extensions of the most promising ones on pretrained models. Namely, given 110 M parameters, BERT’s hidden representations from multiple layers, and many tokens, we try diverse ways to extract optimal sentence embeddings. We test various token aggregation and representation post-processing techniques. We also test multiple ways of using a general Wikitext dataset to complement BERT’s sentence embeddings. All methods are tested on eight Semantic Textual Similarity (STS), six short text clustering, and twelve classification tasks. We also evaluate our representation-shaping techniques on other static models, including random token representations. Proposed representation extraction methods improve the performance on STS and clustering tasks for all models considered. Very high improvements for static token-based models, especially random embeddings for STS tasks, almost reach the performance of BERT-derived representations. Our work shows that the representation-shaping techniques significantly improve sentence embeddings extracted from BERT-based and simple baseline models.

Chloride-salt separation type nanofiltration membranes for efficient crude salt refinement

Application of unrefined crude salt not only leads to a serious equipment scaling and a low product quality, but also increases operational risks and poses health hazards. Therefore, it is essential to refine crude salt prior to its use. However, traditional methods of crude salt refinement are energy-intensive and probably produce additional chemical by-products. In this work, membrane technology was used to accomplish the refinement of crude salt due to its unique characteristics, such as environmentally friendly and lower energy consumption. To improve the separation performance, the membranes were fabricated by simply adjusting the aqueous monomer concentration. For single-salt feed, the membrane (M10) prepared with a high ratio of 10 exhibited excellent salt rejection due to the formation of a thicker and denser PA layer. And it owned a superior MgCl2/NaCl selectivity of ∼22.0, demonstrating that the membrane can achieve outstanding selectivity for mono-/divalent ions. The selectivity of Mg2+/Na+ increased to ∼23.5 for bi-salt feed and further improved to 66.2 for multi-salt feed due to the stronger charge shielding effect. Most importantly, the membrane was also successfully applied in the refinement of crude salt. When using crude salt (NaCl 30 g/L, purity 0.90) as feed, the M10 exhibited both excellent retention of bivalent ions and permeation of univalent ions, resulting in the corresponding Mg2+/Na+ selectivity up to ∼251.6 and the tremendous enhancement in the Na + purity of ∼0.997. This work offers a feasible strategy for crude salt refinement, and could expand to some other potential applications such as resource utilization in the salinization industry and zero-liquid discharge of industrial wastewater.

Research on Pseudo-Label Improvement Based on Probabilistic Uncertainty

Abstract The pedestrian re-identification task explored in this study has strong application scenarios in real life. Generally, in actual scenarios, the pedestrian samples taken by the camera are unlabeled, so pseudo-labeling is used. However, low-quality image feature extraction and imperfect clustering will produce low-quality and incorrect clustering results, which will have a negative impact on the re-identification performance. This article proposes a dual extraction method (PLR). The core idea of this study is to explore the distribution differences and similarities between two data sets through the pseudo-label method, so that the model can learn the distribution in one data set and apply it to another data set and accurately identify the target person to be found. Therefore, this article proposes a dual refinement method based on probability uncertainty. PLR first removes unqualified clustering results through probability uncertainty, and then continues to refine the remaining ones. Experiments have shown that compared to newer methods, PLR is approximately 1% higher in most tasks.

Effect of Nb and Ti additions on microstructure, hardness and wear properties of AlCoCrFeNi high-entropy alloy

The use of Nb and Ti in the development of high-performance alloys is strategically important for technological advancement. The effect of Nb and Ti additions on the microstructure, hardness and tribological properties of the High Entropy Alloy (HEA) AlCoCrFeNi was evaluated. The addition of 0.6mol of Nb led to the formation of Laves phase, resulting in a B2/Laves type hypoeutectic microstructure. Conversely, increasing Ti concentration led to the formation of a dendritic microstructure, with interdendritic region composed of Cr-rich BCC-A2 and Fe-Ti rich Laves phase. The average hardness values obtained were 790 HV, 597 HV and 731 HV for AlCoCrFeNiNb0.6, AlCoCrFeNiTi0.5 and AlCoCrFeNiTi1.2 HEAs, respectively. The high hardness was attributed to the formation of the Laves phase and Cr-rich phases, in addition to solid solution hardening, as confirmed by the increase in the lattice parameter calculated by Rietveld refinement method for both B2 and BCC-A2 phases. While the three compositions exhibited comparable friction coefficient, the alloy containing Nb showed superior wear behavior. This was evidenced by the lower volume of material loss and the lower wear rate, which can be attributed to the fine and homogeneous hypoeutectic microstructure observed in the ingot’s cross section. The wear mechanisms were studied, characteristics features of abrasion and adhesion wear were observed.

Exploring the anti-diabetic activity of 2,6-diamino-4-chloropyrimidinium-hydrogen oxalate conjugates: Synthesis, crystal structure, quantum computational, drug-likeness, molecular docking and dynamics simulation

Inhibiting carbohydrate-hydrolyzing enzyme has been considered as an effective approach for controlling starch digestion and postprandial blood glucose level. The aim of this study, the salt molecule was prepared to identify potentially effective compound against diabetic activity and discover the inhibitory mechanism of the targeted enzyme, which is determined by In-silico techniques. And also perform the quantum chemical calculation to provide insights about the internal motions of the molecule at the atomic level. The 2,6-diamino-4-chloropyrimidinium-hydrogen oxalate (C4H6ClN4+. C2HO4-) was newly synthesized molecule, which is detailly discussed in this present work. The title compound has been confirmed by single-crystal X-ray diffraction studies, and this structure was refined by the least-square refinement method. The crystal compound exhibited the P-1 space group (centrosymmetric) and the following unit cell parameters: a = 5.5778(6)Å, b = 9.6687(10)Å, c = 9.9144(10)Å, α = 116.342(1) °, β = 92.610(1) °, γ = 106.370(1) °, Z = 2. The molecular structure is stabilized by the networks of N–H···O, O–H···O, C–H···O, and N–H···N hydrogen bond interactions. The Hirshfeld surface (HS) analysis and fingerprint plots revealed strong intermolecular interaction between the molecules. The structural optimization, frontier molecular orbital (FMO), molecular reactivity, molecular electrostatic potential (MEP), and mulliken atomic charges were generated by the density functional theory (DFT) calculated at B3LYP/6-311G++ (d,p) theory level. Drug-likeness and physicochemical properties were predicted. The molecular docking analysis was showed a binding energy of −8.96 kcal/mol within the active site of alpha-amylase, which is responsible for anti-diabetic activity. The protein-ligand complex stability was verified using molecular dynamics studies with 100 ns.

Label refinement network from synthetic error augmentation for medical image segmentation

Deep convolutional neural networks for image segmentation do not learn the label structure explicitly and may produce segmentations with an incorrect structure, e.g., with disconnected cylindrical structures in the segmentation of tree-like structures such as airways or blood vessels. In this paper, we propose a novel label refinement method to correct such errors from an initial segmentation, implicitly incorporating information about label structure. This method features two novel parts: (1) a model that generates synthetic structural errors, and (2) a label appearance simulation network that produces segmentations with synthetic errors that are similar in appearance to the real initial segmentations. Using these segmentations with synthetic errors and the original images, the label refinement network is trained to correct errors and improve the initial segmentations. The proposed method is validated on two segmentation tasks: airway segmentation from chest computed tomography (CT) scans and brain vessel segmentation from 3D CT angiography (CTA) images of the brain. In both applications, our method significantly outperformed a standard 3D U-Net, four previous label refinement methods, and a U-Net trained with a loss tailored for tubular structures. Improvements are even larger when additional unlabeled data is used for model training. In an ablation study, we demonstrate the value of the different components of the proposed method.

Annular two-phase flow in a small diameter tube: OpenFOAM simulations with turbulence damping vs optical measurements

In this work, numerical simulations are performed to predict two-phase annular flow of refrigerant R245fa inside a 3.4 mm diameter vertical channel. The VOF (Volume of Fluid) method implemented in an OpenFOAM solver is used to accurately track the vapor-liquid interface. A 2D axisymmetric domain is considered and the Adaptive Mesh Refinement (AMR) method is applied to the cells near the liquid/vapor interface. The Reynolds-Averaged Navier Stokes (RANS) equations are solved and the k-ω SST model is adopted for turbulence modelling in both the liquid and vapor phase. Simulations are used to calculate instantaneous and mean values of the liquid film thickness at mass flux G = 100 kg m-2 s-1 and vapor quality ranging between 0.2 and 0.85. Numerical results are compared against measurements of the liquid film thickness taken during vertical annular downflow. Previous works from the literature and the deviations observed between present numerical and experimental results suggest the need for turbulence damping at the vapor-liquid interface by adding a source term in the ω equation. The simulations show that a low value of the turbulence damping parameter (e.g. 1) causes the average liquid film thickness to increase by 25 %–52 % compared to the non-damped scenario. The interface presents large amplitude disturbance waves in the non-damped case, whereas small ripple waves are predicted when turbulence damping is introduced. Furthermore, the difference between the application of a symmetric and asymmetric treatment for the source term is analysed. From the comparison between experimental data and numerical simulations, it emerges that the value of the correct damping source term to be applied is strictly dependent on the vapor quality.

Matchability and Uncertainty-Aware Iterative Disparity Refinement for Stereo Matching

After significant progress in stereo matching, the pursuit of robust and efficient ill-posed-region disparity refinement methods remains challenging. To further improve the performance of disparity refinement, in this paper, we propose the matchability and uncertainty-aware iterative disparity refinement neural network. Firstly, a new matchability and uncertainty decoder (MUD) is proposed to decode the matchability mask and disparity uncertainties, which are used to evaluate the reliability of feature matching and estimated disparity, thereby reducing the susceptibility to mismatched pixels. Then, based on the proposed MUD, we present two modules: the uncertainty-preferred disparity field initialization (UFI) and the masked hidden state global aggregation (MGA) modules. In the UFI, a multi-disparity window scan-and-select method is employed to provide a further initialized disparity field and more accurate initial disparity. In the MGA, the adaptive masked disparity field hidden state is globally aggregated to extend the propagation range per iteration, improving the refinement efficiency. Finally, the experimental results on public datasets show that the proposed model achieves a reduction up to 17.9% in disparity average error and 16.9% in occluded outlier proportion, respectively, demonstrating its more practical handling of ill-posed regions.

Implicit Is Not Enough: Explicitly Enforcing Anatomical Priors inside Landmark Localization Models.

The task of localizing distinct anatomical structures in medical image data is an essential prerequisite for several medical applications, such as treatment planning in orthodontics, bone-age estimation, or initialization of segmentation methods in automated image analysis tools. Currently, Anatomical Landmark Localization (ALL) is mainly solved by deep-learning methods, which cannot guarantee robust ALL predictions; there may always be outlier predictions that are far from their ground truth locations due to out-of-distribution inputs. However, these localization outliers are detrimental to the performance of subsequent medical applications that rely on ALL results. The current ALL literature relies heavily on implicit anatomical constraints built into the loss function and network architecture to reduce the risk of anatomically infeasible predictions. However, we argue that in medical imaging, where images are generally acquired in a controlled environment, we should use stronger explicit anatomical constraints to reduce the number of outliers as much as possible. Therefore, we propose the end-to-end trainable Global Anatomical Feasibility Filter and Analysis (GAFFA) method, which uses prior anatomical knowledge estimated from data to explicitly enforce anatomical constraints. GAFFA refines the initial localization results of a U-Net by approximately solving a Markov Random Field (MRF) with a single iteration of the sum-product algorithm in a differentiable manner. Our experiments demonstrate that GAFFA outperforms all other landmark refinement methods investigated in our framework. Moreover, we show that GAFFA is more robust to large outliers than state-of-the-art methods on the studied X-ray hand dataset. We further motivate this claim by visualizing the anatomical constraints used in GAFFA as spatial energy heatmaps, which allowed us to find an annotation error in the hand dataset not previously discussed in the literature.

Accuracy Verification of a 2D Adaptive Mesh Refinement Method by the Benchmarks of Lid-Driven Cavity Flows with an Arbitrary Number of Refinements

Accuracy Verification of a 2D Adaptive Mesh Refinement Method by the Benchmarks of Lid-Driven Cavity Flows with an Arbitrary Number of Refinements