Boundary Character Research Articles

GRAB-Net: Graph-based Boundary-aware Network for Medical Point Cloud Segmentation.

Point cloud segmentation is fundamental in many medical applications, such as aneurysm clipping and orthodontic planning. Recent methods mainly focus on designing powerful local feature extractors and generally overlook the segmentation around the boundaries between objects, which is extremely harmful to the clinical practice and degenerates the overall segmentation performance. To remedy this problem, we propose a GRAph-based Boundary-aware Network (GRAB-Net) with three paradigms, Graph-based Boundary-perception Module (GBM), Outer-boundary Context-assignment Module (OCM), and Inner-boundary Feature-rectification Module (IFM), for medical point cloud segmentation. Aiming to improve the segmentation performance around boundaries, GBM is designed to detect boundaries and interchange complementary information inside semantic and boundary features in the graph domain, where semantics-boundary correlations are modelled globally and informative clues are exchanged by graph reasoning. Furthermore, to reduce the context confusion that degenerates the segmentation performance outside the boundaries, OCM is proposed to construct the contextual graph, where dissimilar contexts are assigned to points of different categories guided by geometrical landmarks. In addition, we advance IFM to distinguish ambiguous features inside boundaries in a contrastive manner, where boundary-aware contrast strategies are proposed to facilitate the discriminative representation learning. Extensive experiments on two public datasets, IntrA and 3DTeethSeg, demonstrate the superiority of our method over state-of-the-art methods.

Dependence of intergranular precipitation on grain boundary characteristics in Ni–Cr–Fe–Mo–Cu alloy

Hastelloy G30 is a typical representative of Ni–Cr–Fe–Mo–Cu alloys, which have excellent corrosion resistance and mechanical properties. High alloying-element content induces a risk of intermetallic precipitation, which is poorly eliminated in steel/nickel bimetallic composite manufactured by hot forming or welding. In this study, aged material was observed using transmission electron microscopy, electron backscatter diffraction, and scanning electron microscopy. The results show that the precipitates formed in Hastelloy G30 after aging at 800 °C comprised Cr- and Mo-rich σ phase. More than 95% of Σ1 and Σ3 boundaries were free from σ phase, and about half of the other low-Σ boundaries were immune to the formation of σ phase. All Σ9, Σ27, and random high-angle grain boundaries exhibited low intergranular precipitation resistance of approximately 5%–10%. The (301) planes of the σ phase and the (−220) planes of the matrix showed semicoherent lattice matching. Higher grain-boundary energy and serrated grain boundaries provided opportunities for σ phase precipitation to occur.

MRL-Net: Multi-Scale Representation Learning Network for COVID-19 Lung CT Image Segmentation.

Accuracy segmentation of COVID-19 lesions in lung CT images can aid patient screening and diagnosis. However, the blurred, inconsistent shape and location of the lesion area poses a great challenge to this vision task. To tackle this issue, we propose a multi-scale representation learning network (MRL-Net) that integrates CNN with Transformer via two bridge unit: Dual Multi-interaction Attention (DMA) and Dual Boundary Attention (DBA). First, to obtain multi-scale local detailed feature and global contextual information, we combine low-level geometric information and high-level semantic features extracted by CNN and Transformer, respectively. Secondly, for enhanced feature representation, DMA is proposed to fuse the local detailed feature of CNN and the global context information of Transformer. Finally, DBA makes our network focus on the boundary features of the lesion, further enhancing the representational learning. Amounts of experimental results show that MRL-Net is superior to current state-of-the-art methods and achieves better COVID-19 image segmentation performance.

Boundary-guided and Region-aware Network with Global Scale-adaptive for Accurate Segmentation of Breast Tumors in Ultrasound Images.

Breast ultrasound (BUS) image segmentation is a critical procedure in the diagnosis and quantitative analysis of breast cancer. Most existing methods for BUS image segmentation do not effectively utilize the prior information extracted from the images. In addition, breast tumors have very blurred boundaries, various sizes and irregular shapes, and the images have a lot of noise. Thus, tumor segmentation remains a challenge. In this paper, we propose a BUS image segmentation method using a boundary-guided and region-aware network with global scale-adaptive (BGRA-GSA). Specifically, we first design a global scale-adaptive module (GSAM) to extract features of tumors of different sizes from multiple perspectives. GSAM encodes the features at the top of the network in both channel and spatial dimensions, which can effectively extract multi-scale context and provide global prior information. Moreover, we develop a boundary-guided module (BGM) for fully mining boundary information. BGM guides the decoder to learn the boundary context by explicitly enhancing the extracted boundary features. Simultaneously, we design a region-aware module (RAM) for realizing the cross-fusion of diverse layers of breast tumor diversity features, which can facilitate the network to improve the learning ability of contextual features of tumor regions. These modules enable our BGRA-GSA to capture and integrate rich global multi-scale context, multi-level fine-grained details, and semantic information to facilitate accurate breast tumor segmentation. Finally, the experimental results on three publicly available datasets show that our model achieves highly effective segmentation of breast tumors even with blurred boundaries, various sizes and shapes, and low contrast.

Production simulation and prediction of fractured horizontal well with complex fracture network in shale gas reservoir based on unstructured grid

In order to accurately simulate the productivity variation characteristics of fractured wells with complex fracture network in shale gas reservoir, based on the multiple migration mechanism of shale gas, the micro-seismic data and discrete fracture model were used to characterize the fracture geometry and complex boundary characteristics, and the comprehensive seepage mathematical model of fractured wells with complex fracture network was established based on the dual porosity-discrete fracture model, and the numerical solution was carried out by combining the unstructured grid and the control volume finite element method. The sensitivity analysis of the influence of key parameters such as fracture conductivity, physical property difference in composite area and Langmuir volume on the production performance of fractured horizontal wells is carried out. This study provides theoretical methods and calculation tools for accurate prediction of productivity change and optimization of production system of fractured horizontal wells with complex fracture network in shale gas reservoirs.

MCSF-Net: a multi-scale channel spatial fusion network for real-time polyp segmentation.

Colorectal cancer is a globally prevalent cancer type that necessitates prompt screening. Colonoscopy is the established diagnostic technique for identifying colorectal polyps. However, missed polyp rates remain a concern. Early detection of polyps, while still precancerous, is vital for minimizing cancer-related mortality and economic impact. In the clinical setting, precise segmentation of polyps from colonoscopy images can provide valuable diagnostic and surgical information. Recent advances in computer-aided diagnostic systems, specifically those based on deep learning techniques, have shown promise in improving the detection rates of missed polyps, and thereby assisting gastroenterologists in improving polyp identification. In the present investigation, we introduce MCSF-Net, a real-time automatic segmentation framework that utilizes a multi-scale channel space fusion network. The proposed architecture leverages a multi-scale fusion module in conjunction with spatial and channel attention mechanisms to effectively amalgamate high-dimensional multi-scale features. Additionally, a feature complementation module is employed to extract boundary cues from low-dimensional features, facilitating enhanced representation of low-level features while keeping computational complexity to a minimum. Furthermore, we incorporate shape blocks to facilitate better model supervision for precise identification of boundary features of polyps. Our extensive evaluation of the proposed MCSF-Net on five publicly available benchmark datasets reveals that it outperforms several existing state-of-the-art approaches with respect to different evaluation metrics. The proposed approach runs at an impressive ∼45 FPS, demonstrating notable advantages in terms of scalability and real-time segmentation.

The channeling effect of symmetrical tilt grain boundaries on helium bubbles in tungsten

Helium atoms can migrate easily and aggregate at various crystal defects, such as grain boundaries, which promote the nucleation and growth of helium bubbles, ultimately causing severe radiation damage to nuclear materials. Structure characteristics of grain boundaries with low-angle symmetrical tilt of the [100] axle and their channeling effects on helium bubbles in tungsten were investigated by molecular dynamics simulations. The triaxial variation curves of the helium bubble size in each grain boundary model and the relationship of the helium diffusion coefficient of each grain boundary to their intrinsic crystal structures are analyzed. The results indicate that the helium bubbles grow into one-dimensional nanochannels at the selected grain boundaries, and there is a significant difference in the maximum number of helium bubbles these grain boundaries can accommodate before forming one-dimensional nanochannels. Furthermore, the existence of helium nanochannels along the 〈100〉 edge dislocation line in the low-angle symmetrical tilt grain boundaries is well verified by molecular dynamics simulation of the helium bubble growth and helium diffusion equations at the 〈100〉 edge dislocation line. This suggests that releasing helium outside of the tungsten matrix through the nanochannel structures of low-angle symmetrical tilt grain boundaries may be a potential strategy to address the root causes of bubble nucleation and swelling of tungsten material.

Characterisation of Microstructure and Special Grain Boundaries in LPBF AlSi10Mg Alloy Subjected to the KoBo Extrusion Process

Grain boundary engineering (GBE) enhances the properties of metals by incorporating specific grain boundaries, such as twin boundaries (TB). However, applying conventional GBE to parts produced through additive manufacturing (AM) poses challenges, since it necessitates thermomechanical processing, which is not desirable for near-net-shape parts. This study explores an alternative GBE approach for post-processing bulk additively manufactured aluminium samples (KoBo extrusion), which allows thermo-mechanical treatment in a single operation. The present work was conducted to examine the microstructure evolution and grain boundary character in an additively manufactured AlSi10Mg alloy. Microstructural evolution and grain boundary character were investigated using Electron Back Scattered Diffraction (EBSD) and Transmission Electron Microscopy (TEM). The results show that along with grain refinement, the fraction of Coincidence Site Lattice boundaries was also increased in KoBo post-processed samples. The low-Σ twin boundaries were found to be the most common Coincidence Site Lattice boundaries. On the basis of EBSD analysis, it has been proven that the formation of CSL boundaries is directly related to a dynamic recrystallisation process. The findings show prospects for the possibility of engineering the special grain boundary networks in AM Al–Si alloys, via the KoBo extrusion method. Our results provide the groundwork for devising GBE strategies to produce novel high-performance aluminium alloys.

Topological characteristics of grain boundary networks during severe plastic deformations of copper alloys

The grain boundary character distribution strongly affects the properties of polycrystalline materials. Grain boundaries of similar characters form networks, whose topological invariants can be considered as distribution descriptors. Understanding the evolution of such descriptors during severe plastic deformations (SPD) can elucidate the evolution of properties and underpin the design of processing routes for target behaviour. For topological analysis of grain boundary networks, polycrystalline materials are considered here as polyhedral complexes with grain boundaries classified into two types — low-angle and high-angle. Changes of grain boundary types are calculated using sub-grain rotations, which reflects the physical mechanism of microstructure evolution during SPD. A non-physical approach, by direct conversions of low-angle to high-angle boundaries, is also explored as a reference to demonstrate the impact of the physical constraint imposed by rotations. Reported is the discovery of topological phase transitions in the grain boundary networks which might take place during severe plastic deformations of different copper alloys. Depending on the evolution approach, the transitions correspond to zeros of the Euler characteristic, or of the logarithm of the inverse connectivity, of the grain boundary network. The relations between these transitions and the fraction of high-angle grain boundaries, and between the fraction of high-angle grain boundaries and plastic strain obtainable experimentally, provide new perspectives for grain boundary engineering and network design. Determining the dominant evolution mechanism and critical accumulated strain for a given material and processing route requires further experimental studies of triple junction evolution.

Kinetics of grain boundary migration in nanosized Al polycrystals

Classical theories of grain growth assume that grain boundary curvature and velocity are linearly correlated. However, recent experimental observations seem to imply that the relationship is more complex in polycrystalline materials. Here, we determined the velocity and curvature of approximately 12000 grain boundaries from molecular dynamics simulations of the annealing of randomly textured nanosized polycrystalline Al. The grain boundary kinetics was studied at different scales and compared with theoretical predictions. At the grain level, that is, attributes averaged over all grain boundaries of a grain, the kinetics behavior showed a linear correlation of curvature and velocity. When studying grain-boundary-level behavior (i.e., attributes of individual grain boundaries are considered), we found a clear correlation between the sign of the grain boundary curvature and the direction of migration. Nevertheless, considering all the grain boundaries, the correlation coefficient between the magnitude of velocity and curvature was found to be low (0.34) although a strong linear correlation was observed for a subset of high angle, long-lived, initially large grain boundaries with uniform curvature distribution and similar 5 crystallographic degrees of freedom. Furthermore, some grain boundaries that are close in crystallographic space showed different kinetics. This elucidates the complexity of grain boundary migration in the nanocrystalline ensemble as a result of complex dependence of grain boundary mobility and energy on grain boundary character, and the existence of other driving forces that affect the behavior of grain boundary migration. Among others, defect distributions and variation in stress distribution between neighboring grains are found to influence grain boundary migration.

Real-time and robust feature detection of continuous marker pattern for dense 3-D deformation measurement

Visuotactile sensing technology has received much attention in recent years. This article proposes a feature detection method applicable to visuotactile sensors based on continuous marker patterns (CMP) to measure 3-d deformation. First, we construct the feature model of checkerboard-like corners under contact deformation, and design a novel double-layer circular sampler. Then, we propose the judging criteria and response function of corner features by analysing sampling signals’ amplitude-frequency characteristics and circular cross-correlation behavior. The proposed feature detection algorithm fully considers the boundary characteristics retained by the corners with geometric distortion, thus enabling reliable detection at a low calculation cost. The experimental results show that the proposed method has significant advantages in terms of real-time and robustness. Finally, we have achieved the high-density 3-d contact deformation visualization based on this detection method. This technique is able to clearly record the process of contact deformation, thus enabling inverse sensing of dynamic contact processes.

A hydrogen diffusion model considering grain boundary characters based on crystal plasticity framework

The influence of grain boundaries (GBs) on hydrogen transportation is a crucial factor in understanding the mechanisms of hydrogen embrittlement (HE) in polycrystalline alloys. Crystal plasticity models have the advantages in simulating the effect of grain microstructure on hydrogen diffusion in polycrystals, but they have not yet been coupled with hydrogen diffusion models that can take into account GB characters. In this study, a hydrogen diffusion model that considers the effect of GBs on the transportation of hydrogen was developed to study the local hydrogen diffusion behavior in bi-crystalline and polycrystalline alloys. The simulation results showed the capability of the model in describing the influence of hydrogen on the material deformation. In addition, this study found that the hydrostatic stress determined the hydrogen concentration in the far-GB region, whereas the GB energy played a control role in and near the GB regions. Notably, hydrogen exhibits a preference for localization near high-angle grain boundaries (HAGBs) and triple junctions. In alloys with medium grain size, hydrogen typically had low average concentration with a more uniform distribution. Furthermore, the hydrogen-assisted crack growth mechanism was analyzed as the continuous coalescence of primary and secondary cracks along GBs. This work presents a hydrogen diffusion model that considers GB effects, and the proposed model has advantages in predicting the deformation and fracture of polycrystalline alloys in hydrogen environments.

Significantly enhancing elevated-temperature strength and ductility of a FeMnCoCr high-entropy alloy via grain boundary engineering: Exploring multi-deformation mechanisms

The objective of the study is to elucidate the influence of grain boundary engineering (GBE) on the high-temperature properties of a metastable FeMnCoCr high-entropy alloy (HEA). Using thermo-mechanical processing (TMP), the grain boundary character distribution (GBCD) was optimized, which involved the increase of coincidence site lattice (CSL) special boundaries and multiple twinning events, and breaking of random high-angle grain boundaries (RHAGBs) network. Furthermore, the deformation strain of the alloy became more uniform at high temperature, dynamic recrystallization was suppressed, and a number of deformation mechanisms were operative. The ductility of studied HEA was significantly enhanced by over 200% at 1073 K, resulting in a remarkable 1.35-fold increase.

The Hydraulic and Boundary Characteristics of a Dike Breach Based on Cluster Analysis

It is important to determine the hydraulic boundary eigenvalues of typical embankment breaches before carrying out research on their occurrence mechanisms and assessing their repair technology. However, it is difficult to obtain the hydraulic boundary conditions of the typical levee breaches accurately with minor or incomplete measured data due to the complexity and instability of the levee breach. Based on more than 100 groups of domestic and foreign test data of embankment/earth dam failures, the correlation between the hydraulic boundary eigenvalues of a breach was established based on the cluster analysis approach. Additionally, the missing values were imputed after correlating and fitting. Meanwhile, the hydraulic boundary parameters and the related equations of a generalized typical breach were obtained through the statistical analysis of the probability density of the dimensionless eigenvalues of the breach. The analysis showed that the width of the breach mainly ranges in 20~100 m, while the water head of the breach is 4~12 m, and the velocity of the breach is 2~8 m/s. The distribution probabilities of all them are about 64~71%. The probability density of the width-to-depth ratio and the Froude number of the breach are both subject to normal distribution characteristics. The distribution frequency of the width-to-depth ratio of 3~8 is approximately 55%, and the Froude number of 0.4~0.8 is approximately 60%. These methods and findings might provide valuable support for the statistical research of the boundary and hydraulic characteristics of the breach, and the closure technology of breach.

Strategy for suppressing abnormal grain growth of ZK60 Mg alloy during solution by pre-compression: A quasi-in-situ study

Abnormal grain growth (AGG) easily takes place in Mg alloys during high-temperature solutions, resulting in deterioration of mechanical properties. Hence, the compression prior to solution (pre-compression) was conducted to suppress AGG, and the microstructure evolution as well as suppressing mechanisms was investigated based on quasi-in-situ analysis. After compression along the transverse direction, <11–20>//ED grains preferentially nucleated and rapidly grew up, and the initial <10–10>//ED texture was weakened. Two grain growth modes of heat-induced and strain-induced grain boundary migrations were found. The former was attributed to the high interfacial energy of grain boundaries with large curvature. The latter consumed the adjacent grains with high storage energy, forming abnormal grains with irregular shapes. The compression with a reduction > 6% could obviously suppress AGG. The suppressing effects were mainly attributed to weakening the size advantage of <11–20>//ED grains, increasing nucleation, reducing grain boundary character distribution, and redistributing storage energy distribution. After 12% compression along the transverse direction, 30° misorientation of <11–20>//ED grains and high energy grain boundaries were reduced. The {10–12} tensile twins and {10–15} high index twins induced by compression increased the nucleation of static recrystallization. Beside, compression introduces high-density dislocations, which also contributed to suppressing AGG behavior during solution.

Uncertainty‐aware iterative learning for noisy‐labeled medical image segmentation

AbstractMedical image segmentation from noisy labels is an important task since obtaining high‐quality annotations is extremely difficult and expensive. There are a lot of approaches proposed for such task. However, some issues like the overfitting on noisy annotations, the limited learning of boundary features, and no consideration of the corrupted local pixels are still not solved. Therefore, a novel approach named uncertainty‐aware iterative learning (UaIL) is proposed for medical image segmentation with noisy labels. UaIL iteratively and jointly trains two deep networks using the original images and their argumented ones through a joint loss function including softened label loss, hard label loss and consistency loss, which encourages UaIL to produce segmentations that are robust to the perturbations in arbitrary semantic space. The uncertainty of labels is estimated based on the predictions in iterative learning, then the original labels are refined, which improves the learning of boundary features in segmentation. To avoid overfitting, a stopping strategy is designed based on the dice coefficient in iterative learning. Experiments on two public datasets verify the effectiveness of UaIL under different levels of annotation noise. Especially, when there are serious noises in the labels, the dice achieved by UaIL is 1.43% to 15.03% higher than the competing approaches on the two public datasets. The UaIL is further verified on a private dataset, which shows its ability of applying in the real application with noisy labels.

Effect of Grain Boundary Characters on Precipitation Behavior and Local Deformation Behavior in Al–Mg–Si Alloy

Effect of Grain Boundary Characters on Precipitation Behavior and Local Deformation Behavior in Al–Mg–Si Alloy

Characterization of local deformation around hydrides in Zircaloy-4 using conventional and high angular resolution electron backscatter diffraction

Zircaloy-4 is used as a fuel cladding material for water reactors, as it has good mechanical properties, corrosion resistance, and a low thermal neutron absorption cross section. However, the mechanical performance of Zircaloy-4 can be reduced during service due to hydrogen uptake and hydride formation. These hydrides are brittle, and often reduce the strength and toughness of materials as well as increase susceptibility to delayed hydride cracking (DHC). In this work, large grain Zircaloy-4 with hydrides was prepared and then cross sectioned using cryo-ion beam polishing, using plasma focused ion beam (pFIB) and broad ion beam (BIB) approaches to enable the preparation of a very high quality flat surface with no preferential etching of either the hydride or zirconium metal (typically metallographic polishing preferentially removes hydrides). Conventional and high angular resolution electron backscatter diffraction (EBSD) analysis were then used to explore morphology, deformation fields, and orientation relationships between the zirconium matrix and hydrides. Four maps were collected for analysis which included hydrides near grain boundaries: (a) where the hydride smoothly decorates across two of the connecting boundaries near a triple junction; (b) where the hydride smoothly decorates the boundary; (c) a mixture of smooth decoration of the interface and protrusion into the grains; (d) fine scale hydride that protrudes into one grain. This work highlights that incompatibility of the hydride within the zirconium matrix is strongly linked to the orientation relationship of the hydride and matrix, and the grain boundary character. These results may enable enhanced understanding of the role of hydrides in fracture as well as stress-induced hydride reorientation and DHC susceptibility.

Cryogenic deformation tailors the dislocation accumulation mode to modify grain boundary character distribution of Incoloy 925

Cryogenic deformation was introduced in thermomechanical processing to restrict dislocation mobility and tailor the initial dislocation distribution at the deformation stage, subsequently followed by annealing treatment to alter the grain boundary characteristic distribution with a view to the feasibility of grain boundary engineering in Incoloy 925. The microstructural evolution deformed at cryogenic (77 K) and room temperature (298 K) under various rolling reductions and annealing conditions was systematically studied to optimize the grain boundary network, which is evaluated using multi-source quantitative statistics of the twin related domains. The results showed that cryogenic deformation can adjust the dislocation accumulation mode and plastic deformation mechanism. Geometrically dislocation distributions, reconstructed by HR-EBSD device, with fishbone-like, ribbon-like or lamellar characteristics can be observed for the deformed specimens. During the subsequent annealing, the samples with 5% and 15% pre-strain are incapable of producing the ideal twin related domains, while the sample annealed at 1075°C with 10% pre-strain show the optimization of GBCD with the highest Σ3n (n = 1, 2, and 3) boundary fraction being 69.67%. Compared with the room temperature pre-deformation, cryogenic deformation can effectively reduce the critical temperature of GBE, and also leads to the Σ3n grain boundaries proportion increasing from 31.47% to 67.46% in 10%/1050°C sample. Meanwhile, with the appropriate strain, cryogenic deformation is more conducive to optimizing the GBCD than room-temperature deformation.

Optimization of Grain Boundary Character Distribution of Fe-19Cr-10Mn-1Ni-0.53N High Nitrogen Austenitic Stainless Steel

High nitrogen austenitic stainless steel (HNASS) has outstanding mechanical properties, but in its hot working, welding, and high-temperature use, there will be a precipitation phase, especially nitride precipitation that reduces its intergranular corrosion, stress corrosion, corrosion fatigue, and other grain boundary-related properties. At present, the traditional methods of suppressing precipitation phase precipitation, such as solution treatment and alloying, have drawbacks. Grain boundary character distribution (GBCD) optimization of Fe-19Cr-10Mn-1Ni-0.53N HNASS was characterized by electron backscatter diffraction (EBSD) in this work. This study reveals that after grain boundary engineering (GBE) treatment, the percentage of low Σ coincidence site lattice (CSL) boundaries of the solid solution treated sample rises from 53.94% to 82.41%, because the sample was cold-rolled (CR) by 5%, followed by annealing at 1423 K for 10 min, and the twin related domains (TRD) size increases from 23.99 μm to 169.82 μm and ν (the ratio of TRD size to grain size) raises from 1.74 to 7.52. The connectivity of the random high-angle grain boundary (RHAGB) web is interrupted and the GBCD of the experimental steel is remarkably optimized.