Atomic structures of a Lu-segregated grain boundary (GB) in α-Al2O3 are identified using hybrid Monte Carlo and molecular dynamics (MCMD) simulations based on a neural-network potential (NNP) trained on density-functional-theory (DFT) data, in combination with scanning transmission electron microscopy (STEM). The NNP accurately reproduces the relationship between the potential energy and atomic structures. This enables us to screen candidate atomic structures by performing many structural relaxations and long time-scale MD simulations, prior to final DFT validation, significantly reducing computational cost. The NNP predicts that multiple Lu configurations are energetically favorable, with variations in the occupied site and segregation level. The Lu atomic configurations observed in the experimental STEM images are fully explained by the present calculations, allowing for quantitative analyses of the atomic and electronic structures. The present NNP approach opens the way for a deeper understanding of impurity-segregated GBs at the atomic level.

Thermal flutter and buckling boundaries of PSC structures with surface cracks

Garnet Li7La3Zr2O12 electrolyte is considered a key enabler of solid-state batteries with Li metal electrodes, but the grain boundaries impair its performance. To date, the understanding of grain boundary structures and its impact on performance remains elusive. Here, we show that element segregation at Li7La3Zr2O12 grain boundaries critically governs Li transport and nucleation. During conventional sintering, Al, Ta, and La segregate at grain boundaries, locally depleting Li and creating space-charge layers that lower total ionic conductivity. Simultaneously, this segregation leads to higher electronic conductivity along grain boundaries, which promotes Li nucleation at grain boundary edges with increased risk of dendrite formation. The underlying mechanism of segregation is governed by both thermodynamic driving forces and diffusion kinetics. Building on this understanding, we develop a strategy to achieve segregation-free grain boundaries through a rapid sintering protocol that utilizes the onset of solid-state softening. This approach yields transparent, polycrystalline Li7La3Zr2O12 with negligible grain boundary impedance and enhanced dendrite tolerance. By elucidating the structural origins and electrochemical consequences of grain boundary segregation, this work provides a guidance for the rational optimization of solid electrolytes.

Structural optimization plays a crucial role in enhancing the performance of magnetic actuators. Traditional design approaches, such as parametric scanning, are limited by their reliance on empirical geometries. Meanwhile, widely used topology optimization techniques—including the Solid Isotropic Material with Penalization (SIMP) method and level-set methods—often encounter difficulties such as a large number of design variables, high computational cost, and unclear structural boundaries. To overcome these limitations, this paper proposes a novel gradient-free topology optimization method based on superellipses for designing magnetic actuator yokes. The proposed approach offers three key benefits: (1) It requires very few design variables, with each superellipse described by only seven parameters, thereby reducing the dimensionality of the design space and simplifying the optimization problem. (2) It yields clear and smooth structural boundaries without the need for post-processing. (3) It operates without gradient information, employing stochastic algorithms such as genetic algorithms that rely solely on objective function evaluations. A case study on yoke optimization demonstrates that our method achieves magnetic force output comparable to or better than the SIMP method, but with significantly fewer variables and a simpler implementation. This work provides an efficient and new tool for the conceptual design of magnetic actuators and related electromagnetic devices.

Dissipative dynamics across physical systems exhibit organizing structural boundaries. The dimensionless damping ratio [Formula: see text] defines a stability architecture in which the critical threshold [Formula: see text] marks a second-order non-Hermitian Exceptional Point (EP2). Within spatially flat ΛCDM, an exact algebraic identity is demonstrated linking the onset of cosmic acceleration to critical damping of structure growth: [Formula: see text]. This identity is a structural reformulation of the standard Friedmann and growth equations, not a new physical law; its scientific value lies in recasting a known kinematic transition as a stability-phase transition and in generating the falsifiable prediction that departures from flat ΛCDM produce a measurable, nonzero offset between the [Formula: see text] and [Formula: see text] transitions (identically zero under flat ΛCDM, generically nonzero in extensions). A substrate inheritance relation is then proposed as a leading-order projection approximation, whereby emergent modes acquire effective parameters from substrate precursors. Within this framework, the observed particle distribution is consistent with stability organization: long-lived matter occupies localized stability basins at [Formula: see text], while short-lived resonances occupy a secondary [Formula: see text]-1 band whose sub-cluster evidence is suggestive rather than conclusive. The framework is presented as a structurally grounded classification with defined scope and testable consequences.

Synergistic engineering of A-site defects and grain boundaries in LaCoO3 for efficient catalytic oxidation of chlorobenzene and toluene.

Complementary cross-gated fusion framework for brain tumor segmentation using MR images (CoGFu Net).

Rectovaginal fistula from perineal body injury during delivery is rare, with limited case volumes preventing standardized protocols. We developed an AI system for anatomical landmark detection and evaluated its educational potential during surgery. This pilot feasibility study developed an AI system using HyperSeg semantic segmentation to detect anatomical landmarks (perineal body, vaginal wall, rectum) during rectovaginal fistula repair. Training data comprised 2000 annotated images from 10 cases; 100 images from 5 independent cases served as the test dataset. Dense sampling, reverse-chronological annotation, and multi-structure detection were employed to maximize performance with limited data. Surgical outcomes and intraoperative communication were compared between a pre-AI period and an AI evaluation period (5 cases each; 25 consecutive cases total). The trainee assistant-with no prior experience in this procedure-referenced the AI display while the supervisor guided without viewing it. The AI model achieved Dice coefficients of 0.655 ± 0.185 (perineal body), 0.672 ± 0.167 (vaginal wall), and 0.705 ± 0.144 (rectum). The rectum met the predefined threshold of 0.7; the vaginal wall and perineal body approached but did not reach this threshold, reflecting the inherently ambiguous boundaries of these thin membranous structures. Temporal comparison between periods showed: 221% difference in dissection time efficiency (p = 0.008), 73% reduction in blood loss (p = 0.056), and 185% difference in repair time efficiency (p = 0.056). Despite shorter operative times, intraoperative communication increased substantially: Supervisor instructions increased 71% (dissection phase, p = 0.008) and 68% (repair phase, p = 0.012); assistant questions increased 67% and 100%, respectively (all p ≤ 0.012, Cliff's δ = 1.000). This pilot feasibility study demonstrated technical feasibility of AI landmark detection in rectovaginal fistula repair using only 10 training cases, with the rectum achieving the predefined Dice threshold. A temporal increase in intraoperative communication was observed following AI introduction, suggesting potential facilitation of supervisor-trainee interactions. Future controlled studies are required to validate these findings.

Larvae of the caddisfly Arctopsyche grandis BANKS build protective structures and spin silken capture nets in flowing water. Caddisfly H-fibroin, the major protein component of its silk fibers, has a blocky structure with repeating units defined as beginning with a [(SX)nE]m region followed by a G-rich spacer. Previous observation of H-fibroin allelic variation in haploid-resolved individuals led us to investigate allelic variation within two geographically close but separated natural populations of A. grandis. The genomes of 18 individuals were sequenced, and 34 haploid-resolved H-fibroin sequences were extracted. Twenty-four unique alleles were identified in 18 genomes, revealing the dynamic nature of the H-fibroin gene. H-fibroin length variations of at up to 25% were tolerated. The major source of the length variations were large-scale deletions and insertions of entire [(SX)nE]m blocks. Small scale indel events were numerous, nonrandomly distributed, and constrained to a few types. One, a 44 residue indel comprising two (SX)nE motifs changed m ± 2 by splitting direct tandem repeats without disrupting tertiary structure or block boundaries. The G-rich spacers are of two types, the first distinguished by repeating GLGPH pentapeptides. Indels within this spacer type occur as multiples of the GLGPH pentapeptide. The other category of G-rich spacer was confined to a narrow length distribution. Overall, the results demonstrate the rapid evolution of the caddisfly H-fibroin gene and the wide range of H-fibroin structural polymorphism tolerated in functional capture net silk. At the same time, the limited nature of the indels point to the critical structural features of H-fibroin.

In recent years, artificial intelligence in medicine has evolved from single recognition tasks toward structural understanding, spatial reasoning, and clinical interpretability. High-quality anatomical data have become a key factor in further development. Driven by digital tomography, three-dimensional reconstruction, and multimodal technologies, body-donor-derived specimens and digital anatomical datasets, characterized by clear structural boundaries, stable spatial relationships, and fine-grained detail, are being transformed into computable, annotatable, and reusable digital anatomical resources. These resources are playing an increasingly important role in medical artificial intelligence. This narrative review summarizes the multiple roles of body-donor-derived data in medical AI. They serve as foundational resources that provide high-fidelity training data and fine-grained annotation systems. They also serve as validation references for improving algorithm credibility. In addition, they act as a substrate for AI-driven transformation in data processing, three-dimensional modeling, and intelligent applications in education, clinical practice, and forensic medicine. Their main strengths lie in anatomical authenticity, fine-grained annotatability, and structural validation utility, while their limitations include sample size, the postmortem-in vivo domain gap, annotation cost, and data governance. In the future, body-donor-derived data should become a core foundation for anatomical priors and structural gold standards, and should be deeply integrated with large-scale clinical imaging, multimodal intelligent analysis, and cross-domain learning to support the development of medical AI from high performance toward higher credibility and translational value.

SLAM (Simultaneous Localisation and Mapping) is an important component in robotics, providing a map of an environment and enabling localisation and navigation. While 3D LiDAR odometry and mapping systems have advanced in recent years, producing accurate motion estimates and detailed 3D maps, high-quality 2D occupancy grid maps (OGMs) remain challenging to obtain in large, complex indoor environments. OGMs are often degraded by drifts in odometry, sensor artefacts, and partial observability, resulting in maps with fractured walls, double boundaries, and artefacts that limit readability for mapping-centric tasks such as floor plan creation. To address this, we propose Transformation & Translation Occupancy Grid Mapping (TT-OGM), a system-level pipeline that targets map fidelity. TT-OGM leverages 3D scan registration to stabilise 2D map construction via projection and standard occupancy updates, then applies a learned GAN-based refinement module as post-processing to remove artefacts, regularise structure, and complete small missing regions. To enable training at scale, we introduce an offline DRL-based data generation process that produces paired but weakly aligned erroneous/clean OGMs spanning diverse error modes and severities. We demonstrate TT-OGM in real-time on a building-scale dataset collected at Loughborough University and evaluate map fidelity against a registered floor-plan reference using mIoU, masked SSIM, and occupied-boundary F1. We additionally report localisation accuracy on S3Ev2 using translation ATE (RMSE) against Cartographer and SLAM Toolbox (Karto). Our results show that 3D registration improves baseline 2D map quality over standard 2D SLAM outputs, and that GAN refinement further increases structural consistency and boundary accuracy in our pipeline. Additional ablations on synthetic stress tests and qualitative transfer to unseen Radish sequences show that the refinement module consistently improves OGM readability under common noise, moderate drift, and clutter conditions. • System-level Transformation-Translation (TT) pipeline for generating 2D OGMs from 3D LiDAR registration/odometry. • Learned post-processing module that refines OGMs by reducing common mapping artefacts and improving boundary fidelity. • DRL-driven synthetic data generation producing paired degraded/clean OGMs with controllable error modes for training and stress testing.

This paper presents a comparison of the acoustic absorption performance of micro-perforated panel with different perforation shapes. The acoustic impedance models of different perforation shapes (circular, triangular, and square cross section and variable cross section tapered micro perforations) are established, and the end correction of acoustic impedance for different micro perforation shapes are optimized. Based on the acoustic electric analogy method, the coupled model of flexible micro-perforated panel absorber considering the vibration of the substrate panel is established, and the Rayleigh-Ritz method is employed to obtain the panel's natural frequencies, while the Spectro-Geometric Method is used to construct the panel's displacement function. The acoustic impedance formula for flexible micro-perforated panel is derived by the modal superposition principle. The semi-analytical results are verified against finite element method results, confirming the accuracy of the semi-analytical model. In addition, the effects of structural parameters (perforation shapes, microporous length, cross-sectional area, perforation ratio and back cavity depth) and boundary conditions on the acoustic absorption coefficients are analyzed and summarized. Finally, an impedance tube test system is designed to measure the sound absorption coefficient of different hole shapes. The experimental results aligned well with the semi-analytical model, proving the accuracy of the proposed approach.

A comparative study of deep learning-based zebrafish image segmentation methods.

Construction and applications of iterative methods for finding approximate solutions of nonlinear equations having unknown zeros of multiplicity with fractal geometry and dynamical behavior.

HighlightsWhat are the main findings?To address the distinct geometric characteristics and scale variations in shipwreck targets, SW-Net was proposed as a specialized encoder–decoder architecture that fuses high-level semantic context with fine-grained spatial details through a multi-scale input module and refined skip connections.To better capture complex shapes, a directional filter bank and a directional attention mechanism are introduced. Steerable Gaussian kernels are used to extract structural boundaries, while orientation-specific features are adaptively weighted to reduce the effects of reverberation.What are the implications of the main findings?Embedding geometric constraints and directional priors into lightweight architectures proves more effective than increasing model depth for distinguishing man-made targets from seabed backgrounds.The computational efficiency of SW-Net enables real-time deployment on resource-constrained autonomous underwater vehicles for “search-and-inspect” missions, reducing the labor and costs of large-scale surveys.Side-scan sonar is a critical instrument for underwater cultural heritage preservation, as it allows large-scale detection of shipwrecks in turbid waters where optical methods fail. However, the automated segmentation of these targets remains a significant challenge, as severe speckle noise and complex seabed reverberations often obscure the distinctive geometric features of submerged structures. To address this challenge, this paper proposes SW-Net, which utilizes a multi-scale input strategy and a novel Directional Filter Bank to inject physical priors into the feature extraction process. Furthermore, by coupling this with a directional attention mechanism, the network dynamically modulates structural features to accurately segment targets despite intensity inversions and speckle noise. As demonstrated by the experimental results on the AI4Shipwrecks dataset, the SW-Net outperforms seven representative segmentation architectures, achieving the highest intersection over union of 39.43% and an F1-score of 56.56%. In addition, the model exhibits superior robustness against complex seabed interference while maintaining the lowest computational complexity of 4.01 million parameters among the evaluated methods. Taken together, the SW-Net is proposed to offer a practical solution for shipwreck detection on resource-constrained autonomous underwater vehicles.

Robust and Interpretable AI for Acute Appendicitis: A Simulation-to-Clinical Validation Pipeline.

MotivationCryogenic electron tomography (cryo-ET) enables in situ visualization of macromolecular and cellular structures from tilt-series projections. Reconstruction quality is often compromised by extremely low signal-to-noise ratio (SNR) and vignetting artifacts arising from detector truncation under constrained acquisition geometries. In practice, existing methods frequently struggle to balance noise robustness, computational efficiency, and stability under these conditions.ResultsWe propose a robust, scalable, and parallelizable variational reconstruction framework that integrates a geometrically consistent data fidelity term with an implicit boundary-handling mechanism to mitigate truncation-induced artifacts without volume padding. A composite sparse regularizer integrating anisotropic total variation and curvelet-domain sparsity is employed to preserve structural boundaries and multiscale directional features. The resulting optimization problem is efficiently solved using the primal-dual hybrid gradient (PDHG) algorithm without nested inner iterations, for which we provide rigorous theoretical guarantees of stability and convergence. Experiments on simulated and experimental cryo-ET datasets demonstrate substantial noise suppression and contrast enhancement while preserving fine structural details under realistic, severely noise-limited and truncated acquisition conditions. These improvements lead to enhanced interpretability and facilitate downstream structural analysis, while achieving significantly reduced runtime compared to existing methods at comparable reconstruction quality.Availability and implementationOur code available at https://github.com/icthrm/CSRT. The real datasets used in this study are publicly available from EMPIAR and the Caltech Electron Tomography Database.

Computerized diagnosis of brain tumor using graph based CNN classification.

Quantifying and localizing groundwater discharge is inherently difficult. It requires knowledge about hydraulic conductivity and the hydraulic gradient on the scale of interest. Conventional hydraulic testing, such as pumping tests, may fail in the presence of heterogeneity and complex structural boundaries. While advanced 2D and 3D hydraulic tomography may resolve small-scale heterogeneity, it is typically limited to small spatial scales and requires costly field installations. We propose a simplified tomographic approach using a limited number of pumping and observation wells spatially distributed over a well profile in the order of 100 m transverse to the direction of ambient flow. To infer the spatially variable hydraulic-conductivity field from drawdown data with its uncertainty, we apply an iterative ensemble smoother. Subsequently, the posterior ensemble of hydraulic-conductivity fields is used to calculate total and specific discharge based on the observed ambient hydraulic heads in the same wells. We test our approach in a synthetic scenario mimicking a channel-like aquifer such as the quaternary fill in a small river valley. The results demonstrate that multiple spatially distributed pumping tests are suitable to quantify total discharge and its associated uncertainty. The approach is more reliable than a conventional one that estimates effective transmissivity from fitting analytical solutions to pumping-test data. The tomographic analysis additionally allows locating spatial patterns of specific discharge at a resolution similar to the spacing of the wells, which may be important when assessing and remediating contaminant plumes.

Understanding the influence of processing routes on grain boundary structure and crystallographic texture is essential for optimizing the magnetic performance of nanocrystalline Nd-Fe-B magnets. In this study, a five-parameter crystallographic analysis was employed to systematically investigate the influence of spark plasma sintering and hot deformation sintering on the grain boundary plane distribution and crystallographic texture of nanocrystalline Nd-Fe-B magnets. Its reveals that the hot deformation process confers marked advantages in promoting strong c-axis texture and tailoring grain boundary configurations. In hot deformed samples, [001]-oriented grains predominate, with the c-axis strongly aligned along the pressing direction, resulting in a significantly enhanced texture strength compared to spark plasma sintering samples. The grain boundary plane distribution in hot deformed samples exhibits a pronounced preference for low-index planes such as (001) and (110), with a peak intensity reaching 9.9 multiples of random distribution. In contrast, spark plasma sintering samples display a more randomized boundary plane distribution. Further, the five-parameter analysis indicates that the hot deformation processing promotes stronger preferential grain boundary plane alignments within specific misorientation ranges. Indeed, hot deformed samples demonstrate superior magnetic properties, with a remanence of 12.5 kG and a maximum energy product of 35.7 MGOe. Collectively, these findings clarify the intrinsic distinctions in microstructural evolution between the spark plasma sintering and hot deformation processes and underscore the utility of the five-parameter analysis in linking the processing pathways to microstructure and performance. This work provides a theoretical foundation for the grain boundary engineering and the optimization of processing routes toward high-performance rare earth magnets.