Search Results (16,859)

Aero-engine defects often exhibit micro-scale and high-frequency characteristics under complex metallic textures, which makes precise segmentation difficult. Most existing pixel-level methods rely on spatial-domain modeling and lack frequency-domain decoupling. As a result, high-frequency details are easily hidden by low-frequency background information. In addition, repeated downsampling weakens the representation of fine-grained structures, leading to inaccurate boundary localization and limited robustness. To address these issues, a spectrum-driven hierarchical learning network is proposed for aero-engine defect segmentation. First, a dual-band spectral module is constructed using the discrete cosine transform to separate high-frequency and low-frequency components, providing stable and physically meaningful frequency-domain priors for the network. Second, a detail-guided module is designed where high-frequency features adaptively guide skip connections, compensating information loss during encoding and improving boundary recovery. Furthermore, a low-frequency-driven region-aware modeling module is developed. The internal defect regions, boundary areas, and background regions are modeled hierarchically. A dynamic hyper-kernel generation mechanism performs region-sensitive convolutional modeling, improving adaptation to complex structural variations. Extensive experiments on the Turbo19 and NEU-Seg datasets demonstrate that the proposed method produces accurate defect boundaries and achieves mIoU scores of 89.82% and 91.44%, improving over the second-best method by 5.22% and 4.42%, respectively. Full article

Quadrotor motion planning in cluttered environments presents significant challenges in achieving both computational efficiency and trajectory smoothness, particularly in low-altitude economy and intelligent energy system applications where autonomous aerial vehicles perform infrastructure inspection and power line monitoring. Many existing methods either rely on sampling-based algorithms that suffer from long computation times and suboptimal paths, or employ trajectory representations that produce high-order derivative discontinuities unsuitable for agile flight. In this work, we propose an efficient hierarchical motion planning framework that integrates a JPS–Bresenham-based path search with safe flight corridor construction and Bézier curve optimization. Our approach addresses trajectory generation through a two-stage process: a front-end path search that efficiently identifies collision-free paths with reduced waypoints, followed by a back-end optimization that leverages convex safe corridors with overlapping regions to expand the solution space. Through comprehensive benchmark experiments across six different map scenarios, we demonstrate that our method outperforms RRT* and PRM in both path quality and computational efficiency. Monte Carlo experiments across varying map sizes and obstacle densities confirm robustness and scalability advantages. Comparative studies with state-of-the-art planners demonstrate superior success rates and cost efficiency while maintaining strict kinodynamic feasibility. The Bézier-based optimization reduces snap integral by up to 55% compared to ordinary polynomial approaches, demonstrating its superiority for fast quadrotor trajectory planning in complex environments. Full article

Imaging-based multi-omics derived from digital histopathology provides a valuable approach for characterizing tumor heterogeneity from routine clinical specimens. However, robust multi-cancer histopathological analysis remains challenging due to pronounced intra-tumor variability, inter-organ morphological overlap, and sensitivity to staining and acquisition variations, which can limit the generalizability of deep learning models. These limitations are largely driven by insufficient representation learning, particularly in multi-organ and multi-class diagnostic settings. In this study, we propose a hierarchically regularized representation learning framework for multi-cancer histopathological image analysis that models imaging-based features across multiple organs and diagnostic categories. The framework integrates complementary mechanisms to capture fine-grained cellular morphology, long-range tissue architecture, and organ-aware diagnostic semantics within a unified computational model. A hierarchical supervision strategy guides the network to reduce entanglement between organ-level structural characteristics and disease-specific diagnostic patterns in the learned representations. The method operates without pixel-level annotations or handcrafted morphological priors, supporting scalable experimental evaluation. We demonstrate the approach on balanced lung and colon cancer histopathology cohorts, achieving 96.5% accuracy on lung cancer classification and 96.8% accuracy on colon cancer classification. Ablation and robustness analyses further validate the contributions of hierarchical regularization and consistency learning. Overall, this work provides a demonstrated proof-of-concept framework for representation-centric imaging-based analysis in multi-organ histopathology under the evaluated dataset conditions. Full article

The widespread adoption of encryption protocols such as TLS 1.3 has significantly reduced the visibility of packet payloads, limiting the effectiveness of traditional traffic analysis methods. Recent deep learning approaches attempt to learn representations directly from raw byte sequences; however, in encrypted environments, byte-level patterns often exhibit high entropy and unstable ordering, raising concerns about their reliability. In this work, we revisit the roles of content and structural information in traffic classification and argue that effective modeling should move beyond content-only representations. We propose a structure-aware framework that models hierarchical relationships across fields, layers, and sessions while representing byte information using compact, permutation-invariant summaries. In addition, we introduce a hierarchical shuffle pretraining strategy to capture relational dependencies and an adaptive inter-level gating mechanism to dynamically integrate multi-level representations. Extensive experiments on multiple datasets with varying levels of encryption demonstrate that byte-level sequential patterns are not always essential, while structural information provides consistent complementary cues. Furthermore, the importance of different structural levels varies across datasets, highlighting the need for adaptive multi-level modeling. The proposed method achieves strong performance across diverse datasets, including highly encrypted traffic, while maintaining robustness under domain shifts and limited data scenarios. These results suggest that combining compact content representations with structural context and adaptive integration is a promising direction for encrypted traffic analysis. Full article

As rural revitalization advances, the age-friendliness of public spaces directly impacts the well-being of left-behind elderly populations. However, the spatial and social marginalization of these vulnerable groups in tourism-driven peri-urban villages remains critically underexplored. To bridge this gap, this study proposes a quantitative evaluation framework integrating space syntax and multiple linear regression to investigate the matching mechanism between physical spatial layout and elderly activity needs. Focusing on Shixia Village in Beijing, surveys and satisfaction assessments were conducted with 30 elderly residents (representing a rigorous 27.3% of the permanent population). Space syntax analysis revealed a distinct “core-periphery” spatial differentiation. Despite a moderate spatial intelligibility (0.586), the rapid decay of integration in peripheral clusters acts as the primary physical bottleneck restricting the elderly’s social radius. Furthermore, regression results indicate that public facility accessibility (β = 0.703) and residential environment quality (β = 0.779) are the core positive drivers of satisfaction (p < 0.001). Conversely, road connectivity exhibited an unexpected negative correlation (β = −0.308). This highlights a crucial “double-edged sword” effect: in traditional villages with tourism development, excessive spatial permeability diminishes the elderly’s territorial sense of security due to external traffic interference. Finally, targeted optimization strategies—including traffic-calming interventions and hierarchical node layouts—are proposed, providing an operational evaluation model and design reference for age-friendly environmental construction in similar peri-urban villages. Full article

This study aimed to enhance the accuracy and interpretability of erosion gully classification within black soil regions by focusing on Changxing Township, Xinxing District, Qitaihe City, Heilongjiang Province as the research site. Utilizing RTK (Real-Time Kinematic) surveying technology, three-dimensional topographic data were collected for 139 actively developing erosion gullies. Key morphological parameters—including gully length, depth, gradient, average top width, average bottom width, and slope gradients on both sides—were extracted to construct interactive features. The variable set was refined through correlation analysis and variance inflation factor (VIF) diagnostics to mitigate multicollinearity. A random forest model was employed as the primary classification approach and benchmarked against logistic regression, support vector machines (SVM), decision trees, and backpropagation neural networks. To address class imbalance, a combination of class weighting, Synthetic Minority Over-sampling Technique (SMOTE), and undersampling methods was implemented. Model tuning and interpretability assessments were performed using cross-validation, grid search optimization, and SHapley Additive exPlanations (SHAP) analysis. The findings demonstrate that the random forest model achieved superior overall performance, with test set accuracy, macro-averaged F1 score, and balanced accuracy values of 0.9143, 0.8087, and 0.8427, respectively. Among imbalance handling techniques, class weighting yielded better results compared to oversampling and undersampling. Feature importance and SHAP analyses identified gully length, average crest width, and their interaction with gully depth as the principal determinants influencing gully grade classification. These results elucidate the synergistic developmental dynamics of gully longitudinal extension, vertical deepening, and lateral widening. The proposed methodology offers valuable technical support for the rapid surveying, classification, and management decision-making processes related to black soil erosion gullies. Full article

Environmental, Social, and Governance (ESG) text classification is important for applications in sustainable finance. However, it remains a challenging task due to domain terminology and regulatory constraints. While transformer-based models achieve strong predictive performance, they often lead to high energy costs and provide limited interpretability. To address these limitations, we introduce ESG-Graph, a lightweight and interpretable graph-based framework for modeling ESG disclosures. In our approach, each sentence is represented as a token-level dependency graph augmented with virtual nodes initialized from a European Sustainability Reporting Standards (ESRS)-based taxonomy, enabling the addition of new ESG concepts without retraining. A multi-layer Graph Attention Network is used instead of transformer encoders, allowing grammatical structure and domain semantics to be modeled jointly. Experiments on three ESG benchmark datasets show that ESG-Graph achieves performance comparable to efficient transformer baselines while consuming up to 60× less energy and using 10× fewer parameters. Additional attribution and ablation studies suggest the method’s policy alignment, interpretability, and robustness. Full article

Background/Objectives: The interplay between cardiovascular risk factors and brain aging remains incompletely understood. We aimed to investigate the comparative associations of coronary artery calcium (CAC) and low-density lipoprotein cholesterol (LDL-C) with MRI-derived volumetric measures of the brain. Methods: In this retrospective, single-center, cross-sectional study, 84 participants who underwent coronary computed tomography for CAC scoring and brain magnetic resonance imaging within 90 days were included; LDL-C levels were available in 69 participants for LDL-based analyses. Brain volumetric measures were obtained using the automated lesionBrain pipeline within the volBrain platform, which performs fully automated tissue segmentation and lesion quantification based on multi-atlas and patch-based approaches. Associations were evaluated using Spearman’s correlation with false discovery rate correction and hierarchical multivariable regression, supported by bootstrap validation and post hoc power analysis. The cohort had a mean age of 58.0 ± 13.0 years (range 19–78) and was derived from routine clinical imaging. Results: LDL-C was positively associated with abnormal white matter volume (ρ = 0.334, p = 0.005), although this did not remain statistically significant after FDR correction (pFDR = 0.090). In fully adjusted models, LDL-C remained the only independent predictor (β = 0.006, 95% CI: 0.002–0.010, p = 0.007; standardized β = 0.225; partial R² = 11.7%), corresponding to a 6.2% increase in abnormal white matter volume per 10 mg/dL increase (derived from log-transformed models). CAC showed only a marginal association (p = 0.059). Post hoc power analysis demonstrated adequate power for LDL-C but insufficient power for CAC. Neither marker was associated with gray matter volume. Conclusions: In this cross-sectional cohort, higher LDL-C was independently associated with greater abnormal white matter volume after adjustment for cardiovascular risk factors, statin use, and CAC. No CAC–brain association was detected in this cohort, but limited statistical power means that small CAC effects cannot be excluded. These findings should be interpreted as associative rather than causal or mechanistic. Full article

White clover (Trifolium repens L.) is a major legume in Mediterranean agroecosystems. This study systematically evaluates 15 autochthonous white clover populations from the Trikala region of Greece, focusing on chemical composition and derived nutritional indices relevant for germplasm characterization and breeding. Fifteen local populations were evaluated under controlled pot cultivation over two consecutive years. Clonal plants were harvested at the early flowering stage. Key traits—crude protein (CP), Ash, Fat, crude fibre (FIBRE), acid detergent fibre (ADF), neutral detergent fibre (NDF), digestible dry matter (DDM), dry matter intake (DMI), and relative feed value (RFV)—were measured. Combined ANOVA revealed significant differences among populations for all traits (p ≤ 0.001), while genotype × year interactions were present but generally minor compared to genotypic effects. Broad-sense heritability was high across most traits (H² = 90.8–99.4%), demonstrating strong genetic control. CP showed positive correlations with DDM, DMI, and RFV, whereas ADF and NDF were negatively correlated with intake and digestibility. Canonical and discriminant analyses showed that a reduced set of traits (CP, Ash, FIBRE, RFV) contributed strongly to differentiation among populations. Hierarchical clustering (heatmap) confirmed these groupings based on fibre and digestibility-related traits. Populations such as Dendrochori and Gorgogyri consistently showed favorable chemical and nutritional profiles, while Fiki and Dendrochori showed the highest stability across years. The present study highlights substantial genetic variability among local white clover populations and identifies trait structures of relevance for germplasm characterization. These findings enhance the characterization of genetic diversity in Trifolium repens and support its potential use in future breeding research under Mediterranean environments. Full article

Series-connected lithium-ion battery packs are widely used in electric vehicles (EVs). However, inevitable inconsistency among cells can cause charge imbalance, accelerated aging, and reduced system safety. To improve the consistency of series-connected battery packs under complex EV operating conditions, this study proposes a hierarchical active balancing system. Bidirectional Buck–Boost converters are employed for intra-group balancing, and distributed flyback converters are used for inter-group balancing. A multi-stage coordinated balancing control strategy is further developed to reduce control complexity and improve balancing efficiency. A 16-cell series-connected battery pack model is established in MATLAB R2024a /Simulink and evaluated under resting, charging, and discharging conditions. The results show that, compared with the conventional single-layer Buck–Boost balancing topology, the proposed method reduces the balancing time by 58.09%, 57.97%, and 58.06%, respectively. These results indicate that the proposed system can effectively improve the consistency and balancing performance of series-connected battery packs, providing a scalable solution for EV battery management systems. Full article

Structured radiology reporting can improve clinical decision support by standardizing clinical findings into hierarchical formats. However, thousands of questions in structured report templates about clinical findings are prohibitively time-consuming, which can limit clinical adoption. Furthermore, early medical VQA datasets primarily focused on free-text and independent question–answer pairs while a recent dataset, Rad-ReStruct, introduced a hierarchical VQA, but the accompanying model still relies heavily on flattened embedding representations and single-path text–image fusion mechanisms that inadequately handle complex hierarchical dependencies in responses. In this paper, we propose DPA-HiVQA (Dual-Path Cross-Attention for Hierarchical VQA), addressing these limitations through two key contributions: (1) multi-scale image embedding representing global semantic embeddings with patch-level spatial features from domain-specific BioViL encoder; (2) dual-path cross-attention mechanism enabling simultaneous holistic semantic understanding and fine-grained spatial reasoning. Evaluated on the Rad-ReStruct benchmark, the model substantially outperforms the established benchmark baseline with an overall F1-score and Level 3 F1-score improvement by 21.2% and 31.9%, respectively. The proposed model demonstrates that dual-path cross-attention architectures can effectively connect holistic semantic understanding and fine-grained spatial detail, paving the way for practical AI-assisted structured reporting systems that reduce radiologist burden while maintaining diagnostic accuracy. Full article

Virtual coupling control of trains is a promising technology for improving railway capacity and operational efficiency. However, existing multi-agent reinforcement learning (MARL) approaches struggle to capture long-sequence temporal dependencies among train states in complex multi-train interaction scenarios, resulting in limited robustness and coordination stability. To address this issue, this paper proposes a Predictive Mamba-based Multi-Agent Soft Actor–Critic (PM-MASAC) framework. A Mamba-based state prediction module is embedded into the centralized Critic network to model historical state sequences and generate predictive state representations, thereby enhancing value estimation accuracy. In addition, a multi-agent aggregated prioritized experience replay (PER) mechanism is introduced to improve the utilization of critical cooperative samples and stabilize training. A hierarchical local–global reward structure is further designed to ensure individual tracking performance while promoting overall formation coordination. Experimental results under realistic railway operating conditions demonstrate that PM-MASAC achieves superior robustness compared with baseline MARL methods. Velocity and spacing tracking errors are maintained within 3% and 1%, respectively, and the steady-state formation success rate exceeds 95.7% in the training environment. Full article

Background/Objectives: Zebrafish fin regeneration serves as a classic model for investigating vertebrate tissue regeneration, yet the core regulatory networks and their crosstalk with the immune microenvironment remain incompletely characterized. This study aimed to identify hub genes, and elucidate the underlying molecular mechanisms and immune microenvironment dynamics during zebrafish fin regeneration. Methods: We integrated multiple bulk RNA-seq datasets of zebrafish fin regeneration from the GEO database, followed by data standardization with batch effect removal. Hub genes were screened via differential expression analysis, weighted gene co-expression network analysis (WGCNA), and predictive models constructed with 13 classic machine learning algorithms. Functional enrichment, time-ordered gene co-expression network (TO-GCN) method, immune infiltration analyses and RT-qPCR validation were further performed. Results: We identified upregulated differentially expressed genes, regeneration-correlated gene modules and their overlapping genes, including 82 candidate genes and 10 hub genes enriched in cytoskeleton remodeling, extracellular matrix organization, and focal adhesion. Temporal analysis uncovered hierarchical gene regulation and functional switching during regeneration. Hub gene expression was significantly correlated with the infiltration of B cells, M1/M2 macrophages and CD8+ T cells, revealing a stage-specific immune microenvironment. RT-qPCR validation showed high consistency with the multi-omics data. Conclusions: This study provides potential gene targets for understanding zebrafish fin regeneration, and offers a valuable reference for investigating the crosstalk between regulatory networks and the immune microenvironment in vertebrate tissue regeneration. Full article

This article examines the concept of the brain (mastiṣka) within the Indian intellectual tradition, tracing its development from the magico-religious medicine of the Atharvaveda (c. 1200–900 BCE) through the classical Āyurvedic texts—the Suśrutasaṃhitā, the Caraksaṃhitā, the Aṣṭāṅgahṛdayasaṃhitā, and the relatively neglected Bhelasaṃhitā—to the subtle-body physiology of Haṭha Yoga literature. Against the background of a comparative analysis with the brain–heart debate in ancient Greek medicine, the article argues that Indian medicine developed a distinctive ‘relational organ model’ in which brain and heart constitute complementary poles of a single vital-cognitive network mediated by the nāḍī (neural-energetic channel) system. This model is neither simply cardiocentric nor encephalocentrist but integrates both within a hierarchical framework. The Bhelasaṃhitā’s unique near-encephalocentrist statement (śiras tālvantare cetanādhiṣṭhānam) reveals a genuine internal debate within classical Indian medicine, while the Haṭhayogic synthesis—locating the ultimate seat of consciousness in the cranial Sahasrāra while preserving the heart as the integrative hub of all channels—represents a coherent integration of both tendencies. The Sāṃkhya philosophical framework provides the metaphysical key to this integration, distinguishing non-material consciousness (puruṣa) from the material cognitive apparatus (antaḥkaraṇa). The article brings into dialogue these historical findings with recent research in neurocardiology, neuroimaging, and prāṇāyāma science to illuminate areas of empirical convergence, contributing to the interdisciplinary dialogue among science, religion, and health on the nature of human flourishing. Full article

Layered double hydroxides (LDH) demonstrate significant potential in flexible superca-pacitors due to their high energy storage capability and adjustable architectures. Never-theless, the practical specific capacitance exhibited by current LDH remains below expec-tations, which is attributed to suboptimal electrode performance and limited active sites. Herein, a novel Fe-doped NiFeCo-LDH/NiFeCoO nanosheet composite supported on car-bon cloth was designed and fabricated as a flexible electrode. In this composite, the Ni-FeCo-LDH supplies numerous reactive centers and accelerates electrochemical kinetics, while the NiFeCoO and carbon cloth significantly improve electrical conductivity and cy-cling stability. Moreover, the heterointerface formed between the LDH and the metal oxide phase further facilitates charge transfer. Owing to such synergistic interactions, the pre-pared NiFeCo-LDH/NiFeCoO@CC electrode demonstrates an excellent areal specific ca-pacitance of 3.282 F cm⁻² at a current density of 1 mA cm⁻², while maintaining a high ca-pacity preservation reaching 88.09% following 5000 cycles. Furthermore, the assembled NiFeCo-LDH/NiFeCoO@CC//AC asymmetric supercapacitor delivers an outstanding en-ergy density reaching 0.302 mWh cm⁻² under a power density of 0.776 mW cm⁻², coupled with an excellent capacitance preservation of 85.29% over 5000 cycles. Meanwhile, it can maintain its initial capacitance under varying bending degrees, rendering it widely ap-plicable for future advanced flexible and wearable electronic devices. Full article