Search Results (22,841)

Pulses (edible dry seeds from legumes) are among the most important crops worldwide. These legumes contain a diverse range of carbohydrates, some of which, such as RFOs (raffinose family oligosaccharides), are considered antinutritional factors due to their negative impact on digestion. An analytical method based on high-power ultrasound-assisted extraction and HPLC analysis was developed and validated for the quantitative determination of soluble carbohydrates (verbascose, stachyose, raffinose, sucrose, galactinol, glucose, galactose, fructose, and myo-inositol) in common beans (Phaseolus vulgaris) and peas (Pisum sativum). The proposed method is fast (extraction time: 1 min), reproducible (RDS: 6.9%), accurate (97.5%), and environmentally sustainable. The method was applied to local collections of P. vulgaris (n = 12) and P. sativum (n = 34), revealing similar qualitative profiles but notable quantitative differences. In P. vulgaris, sucrose and stachyose were predominant, while in P. sativum, verbascose stood out. The total sugar content was higher in peas, especially in commercial varieties, which also showed elevated sucrose levels. Some local varieties combined high sugar content with favorable relative levels between RFOs and other sugars, making them valuable candidates for breeding programs. Linear discriminant analysis enabled classification and prediction of species and varieties, confirming the usefulness of soluble carbohydrates as tools for characterizing these plant materials. Full article

This study develops a comprehensive workflow integrating Headspace Solid-Phase Microextraction Gas Chromatography–Mass Spectrometry (HS-SPME-GC-MS) with advanced supervised machine learning to authenticate the botanical origin of honeys from five distinct floral sources—coriander, orange blossom, astragalus, rosemary, and chehelgiah. While HS-SPME-GC-MS combined with traditional chemometrics (e.g., PCA, LDA, OPLS-DA) is well-established for honey discrimination, the application and direct comparison of Random Forest (RF), eXtreme Gradient Boosting (XGBoost), and Neural Network (NN) models represent a significant advancement in multiclass prediction accuracy and model robustness. A total of 57 honey samples were analyzed to generate detailed volatile organic compound (VOC) profiles. Key chemotaxonomic markers were identified: anethole in coriander and chehelgiah, thymoquinone in astragalus, p-menth-8-en-1-ol in orange blossom, and dill ester (3,6-dimethyl-2,3,3a,4,5,7a-hexahydrobenzofuran) in rosemary. Principal component analysis (PCA) revealed clear separation across botanical classes (PC1: 49.8%; PC2: 22.6%). Three classification models—RF, XGBoost, and NN—were trained on standardized, stratified data. The NN model achieved the highest accuracy (90.32%), followed by XGBoost (86.69%) and RF (83.47%), with superior per-class F1-scores and near-perfect specificity (>0.95). Confusion matrices confirmed minimal misclassification, particularly in the NN model. This work establishes HS-SPME-GC-MS coupled with deep learning as a rapid, sensitive, and reliable tool for multiclass honey botanical authentication, offering strong potential for real-time quality control, fraud detection, and premium market certification. Full article

As people spend more time indoors, the impact of the built environment on psychological health has attracted growing attention. Yet existing studies often have difficulty capturing decision-makers’ reference dependence and loss aversion under uncertainty. To bridge this gap, we propose an evaluation framework comprising three first-level criteria—Outdoor Environment, Physical Comfort (including thermal, lighting, and color environments), and Acoustic Comfort—and determine combined weights by integrating subjective analytic hierarchy process (AHP) judgments with objective entropy weighting based on triangular fuzzy numbers (TFNs). We further incorporate prospect–regret theory to represent loss aversion, expectation-based reference points, and counterfactual regret/rejoicing, and couple it with the VIKOR compromise ranking method, forming an integrated “TFN + Prospect–Regret + VIKOR” approach. The proposed method is applied to four retrofit alternatives for an open-plan office floor (approximately 1200 m²), each emphasizing outdoor environment, physical comfort, acoustic comfort, or no single priority. Experts assessed the schemes using fuzzy linguistic variables. The results show that lighting conditions, thermal comfort, color scheme, and internal noise control receive the highest comprehensive weights. Extensive sensitivity analyses across value/weighting functions and regret-aversion parameters indicate that the ranking of alternatives remains stable while exhibiting clearer separation. Comparative analyses further suggest that, although the overall ordering is consistent with baseline methods, the proposed model increases score dispersion and improves discriminative power. Overall, by explicitly accounting for decision-makers’ psychological behavior and information uncertainty, the framework enables robust and interpretable selection of retrofit schemes for existing office spaces. Full article

The Weishan rare earth element (REE) deposit, located in western Shandong, North China Block, is a typical carbonatite REE deposit and constitutes the third largest light REE resource in China. Its mineralization is closely related to the multi-stage evolution of a carbonatite magma–hydrothermal system. However, the mechanisms governing REE enrichment, migration, and precipitation remain insufficiently constrained from a mineralogical perspective, which hampers the understanding of the ore-forming processes and the establishment of predictive exploration models. Apatite is a pervasively developed REE phase in the Weishan deposit which occurs in multiple generations, and thus represents an ideal recorder of the magmatic–hydrothermal evolution. In this study, different generations of apatite hosted in carbonatite orebodies from the Weishan deposit were investigated using cathodoluminescence (CL), electron probe microanalysis (EPMA), and in situ LA-ICP-MS trace element analysis. Three types of apatite were identified. In paragenetic sequence, Ap-1 occurs as polycrystalline aggregates coexisting with calcite, is enriched in Na, Sr, and LREEs, and shows high (La/Yb)_N ratios, suggesting crystallization from an evolved carbonatite magma. Ap-2 and Ap-3 display typical replacement textures: both contain abundant dissolution pits and dissolution channels within the grains, which are filled by secondary minerals such as monazite and ancylite, and thus exhibit characteristic features of fluid-mediated dissolution–reprecipitation during the hydrothermal stage. Ap-2 is commonly associated with barite and strontianite, whereas Ap-3 is associated with pyrite and monazite and is characterized by relatively sharp grain boundaries with adjacent minerals. From Ap-1 to Ap-3, total REE contents decrease systematically, whereas Na, Sr, and P contents increase. All three apatite types lack Eu anomalies but display positive Ce anomalies. Discrimination diagrams involving LREE-Sr/Y and log(Ce)-log(Eu/Y) indicate that apatite in the Weishan REE deposit formed during the magmatic to hydrothermal evolution of a carbonatite, and that the dissolution of early magmatic apatite, followed by element remobilization and mineral reprecipitation, effectively records the progressive evolution of the ore-forming fluid. Full article

This study investigates the hydrography and geochemical signature in Ilha Grande Bay (RJ, Brazil), focusing on the seasonal intrusion of South Atlantic Central Water (SACW) and its interaction with lead sources. CTD (Conductivity, Temperature, and Depth) data revealed the presence of SACW during the summer campaigns (Mangaratiba/2011 and Frade/2012), characterized by temperatures below 20 °C and salinity between 34.6 and 36. The intrusion is driven by northeasterly winds that favor coastal upwelling, establishing a classic thermohaline stratification. The winter campaigns did not detect SACW, confirming its seasonal nature. Isotopic analysis of Pb in sediments identified six Pb²⁰⁶/Pb²⁰⁷ intervals, indicating multiple sources, including natural contributions, industrial waste, and urban effluents. The Pb²⁰⁶/Pb²⁰⁷ ranges were defined based on cluster analysis and frequency histograms, which are common methods in isotopic provenance studies. An overlap between the most radiogenic isotopic signatures and the presence of SACW suggests that this water mass acts as a vector for transporting trace elements from the deep oceanic region to the coast. This study provides the first evidence that the South Atlantic Central Water (SACW) acts as a seasonal vector, importing a distinct radiogenic Pb isotopic signature onto the continental shelf of Ilha Grande Bay. By synoptically coupling physical water-mass analysis (CTD) with Pb isotopic tracers, we introduce a novel approach that successfully discriminates oceanic from anthropogenic Pb sources, offering a new framework for understanding contaminant transport in coastal areas influenced by boundary currents. It is concluded that the coastal dynamics in Ilha Grande Bay are governed by the seasonal interaction of coastal, continental, and oceanic waters, and that the integration of physical and geochemical data is crucial for understanding mixing processes and contaminant transport in this complex environment. Full article

High-quality code representations are fundamental to code intelligence. Achieving such representations with parameter-efficient fine-tuning (PEFT) remains a key challenge. While code pre-trained models (CodePTMs) offer a robust foundation for general-purpose embeddings, current PEFT approaches face two main obstacles when adapting them: (i) they fail to adequately capture the deep structural characteristics of programs, and (ii) they are limited by the model’s finite internal parameters, restricting their ability to overcome inherent knowledge bottlenecks. To address these challenges, we introduce a parameter-efficient code representation learning framework that combines retrieval augmentation with structure-aware priors. Our framework features three complementary, lightweight modules: first, a structure–semantic dual-channel retrieval mechanism that infuses high-quality external code knowledge as non-parametric memory to alleviate the knowledge bottleneck; second, a graph relative bias module that strengthens the attention mechanism’s capacity to model structural relationships within programs; and third, a span-discriminative contrastive objective that sharpens the distinctiveness and boundary clarity of span-level representations. Extensive experiments on three benchmarks spanning six programming languages show that our method consistently outperforms state-of-the-art parameter-efficient baselines. Notably, on structure-sensitive tasks using the PLBART backbone, RS-Rep surpasses full fine-tuning, delivering a 22.1% improvement in Exact Match for code generation and a 4.4% increase in BLEU scores for code refinement, all while utilizing only about 5% of the trainable parameters. Full article

Accurate indoor positioning and navigation remain significant challenges, with audio sensor-based sound source localization emerging as a promising sensing modality. Conventional methods, often reliant on multi-channel processing or time-delay estimation techniques such as Generalized Cross-Correlation, encounter difficulties regarding computational complexity, hardware synchronization, and reverberant environments where time difference in arrival cues are masked. While machine learning approaches have shown potential, their performance depends heavily on the discriminative power of input features. This paper proposes a novel feature extraction method named Short-Time Homomorphic Deconvolution, which transforms multi-channel audio signals into a 2D Time × Time-of-Flight representation. Unlike prior 1D methods, this feature effectively captures the temporal evolution and stability of time-of-flight differences between microphone pairs, offering a rich and robust input for deep learning models. We validate this feature using a lightweight Convolutional Neural Network integrated with a dual-stage channel attention mechanism, designed to prioritize reliable spatial cues. The system was trained on a large-scale dataset generated via simulations and rigorously tested using real-world data acquired in an ISO-certified anechoic chamber. Experimental results demonstrate that the proposed model achieves precise Direction of Arrival estimation with a Mean Absolute Error of 1.99 degrees in real-world scenarios. Notably, the system exhibits remarkable consistency between simulation and physical experiments, proving its effectiveness for robust indoor navigation and positioning systems. Full article

Accurate and stable mapping of crop types is fundamental to agricultural monitoring and food security. However, inter-annual phenological shifts driven by variations in air temperature, precipitation, and sowing dates introduce systematic changes in the spectral distributions associated with the same day of year (DOY). As a result, the “date–spectrum–class” mapping learned during training can become misaligned when applied to a new year, leading to increased misclassification and unstable performance. To tackle this problem, we develop TABS-Net (Temporal–Spectral Attentive Block with Space–Time Fusion Network). The core contributions of this study are summarized as follows: (1) we propose an end-to-end 3D CNN framework to jointly model spatial, temporal, and spectral information; (2) we design and embed CBAM3D modules into the backbone to emphasize informative bands and key time windows; and (3) we introduce DOY positional encoding and temporal jitter during training to explicitly align seasonal timing and simulate phenological shifts, thereby enhancing cross-year robustness. We conduct a comprehensive evaluation on a Cropland Data Layer (CDL) subset. Within a single year, TABS-Net delivers higher and more balanced overall accuracy, Macro-F1, and mIoU than strong baselines, including 2D stacking, 1D temporal convolution/LSTM, and transformer models. In cross-year experiments, we quantify temporal stability using inter-annual robustness (IAR); with both DOY encoding and temporal jitter enabled, the model attains IAR values close to one for major crop classes, effectively compensating for phenological misalignment and inter-annual variability. Ablation studies show that DOY encoding and temporal jitter are the primary contributors to improved inter-annual consistency, while CBAM3D reduces crop–crop and crop–background confusion by focusing on discriminative spectral regions such as the red-edge and near-infrared bands and on key growth stages. Overall, TABS-Net combines higher accuracy with stronger robustness across multiple years, offering a scalable and transferable solution for large-area, multi-year remote sensing crop mapping. Full article

To address the challenges of high fine-grained similarity and background interference in recognizing teacher teaching behaviors (TTB), this paper proposes a multi-modal network, TBRNet, aiming to improve recognition performance and facilitate teaching reflection. TBRNet leverages CLIP as a semantic prior and introduces two key mechanisms: the Cascaded Collaborative Attention (CCA) module and the Dynamic Query-Driven (DQD). The CCA module performs bidirectional fusion of temporal and semantic features to capture the temporal contextual information of teaching behaviors; the DQD mechanism, using gated semantic prototypes, adaptively focuses on key discriminative regions, improving the model’s ability to distinguish subtle behavioral differences. On the specialized TBU dataset, TBRNet outperforms all baseline models, achieving a Top-1 accuracy of 86.4%. On the public benchmarks UCF-101 and HMDB-51, TBRNet achieves remarkable accuracy rates of 95.8% and 81.7%, respectively, which validates its strong generalization capability across different datasets. This study provides an effective method for efficiently identifying teacher teaching behaviors in real classroom environments. Full article

Within the traditional electronic neural network framework, Generative Adversarial Networks (GANs) have achieved extensive applications across multiple domains, including image synthesis, style transfer and data augmentation. Recently, several studies have explored the use of optical neural networks represented by the diffractive deep neural network (D²NN) for GANs. However, most of these focus on applications of the generative network, and there is currently no well-established D²NN architecture that simultaneously implements generative adversarial functionality. Here, we propose a novel implementation scheme for generative adversarial networks based on all-optical diffraction layers, demonstrating a complete all-optical adversarial architecture that simultaneously realizes both the generative network and the adversarial network (D²NN-GAN). We validated this method on the MNIST handwritten digit dataset, achieving Nash equilibrium convergence with the discriminator accuracy stabilizing around 50%. Concurrently, the average SSIM parameter of generated images reached 0.9573, indicating that the generated samples possess high quality and closely resemble real samples. Furthermore, we extended the framework to the KTH human action dataset, successfully reconstructing the “running” action with a discriminator accuracy of approximately 75%. The D²NN-GAN architecture introduces a fully optical generative adversarial model, providing a practical path for future optical modeling methods, such as image generation and video synthesis. Full article

Background and Objectives: Early recognition of life-threatening organ dysfunction is central to modern sepsis frameworks. We systematically reviewed the prognostic performance and clinical utility of the Neonatal Sequential Organ Failure Assessment (nSOFA) for mortality and major morbidity in NICU populations. The search identified 939 records across databases; after screening and full-text assessment, 16 studies met the inclusion criteria. Methods: Following PRISMA guidance, we searched major databases (2019–2025) for observational or interventional studies reporting discrimination or risk stratification using nSOFA in neonates. Populations included suspected/proven infection and condition-specific cohorts. Heterogeneity in timing, thresholds, and outcomes precluded meta-analysis. Results: A cumulative sample exceeding 25,000 neonates was identified across late- and early-onset infection, all-NICU admissions, necrotizing enterocolitis, respiratory distress, and very preterm screening cohorts. Across settings and timepoints, nSOFA demonstrated consistent, good-to-excellent mortality discrimination, with reported AUROCs ≥ 0.80 and upper ranges near 0.90–0.92; serial scoring within the first 6–12 h generally improved risk classification. Disease-specific applications (NEC, early-onset infection) showed similar discrimination for death or composite adverse outcomes. Conclusions: Evidence from diverse NICU contexts indicates that nSOFA is a pragmatic, EHR-ready organ dysfunction score with robust discrimination for mortality and serious morbidity, supporting routine, serial use for risk stratification and standardized endpoints in neonatal sepsis pathways, aligned with contemporary organ dysfunction–based pediatric criteria. Full article

Background/Objectives: Insulin resistance and ambulatory blood pressure monitoring (ABPM) abnormalities represent distinct but interrelated pathways contributing to cardiovascular risk. The triglyceride–glucose (TyG) index reflects metabolic burden, whereas arterial load—captured through arterial stiffness, blood pressure variability, and morning surge—reflects hemodynamic instability. Whether the coexistence of these domains identifies a particularly high-risk ambulatory phenotype remains unclear. To evaluate the independent and combined effects of metabolic burden (TyG) and arterial load on circadian blood pressure pattern and short-term systolic blood pressure variability. Methods: This retrospective cross-sectional study included 294 adults who underwent 24 h ABPM. Arterial load was defined using three ABPM-derived indices (high AASI, high SBP-ARV, high morning surge). High metabolic burden was defined as TyG in the upper quartile. The “double-high” phenotype was classified as high TyG plus high arterial load. Primary and secondary outcomes were non-dipping pattern and high SBP variability. Multivariable logistic regression and Firth penalized models were used to assess independent associations. Predictive performance was evaluated using ROC analysis. Results: The double-high phenotype (n = 15) demonstrated significantly higher nighttime SBP, reduced nocturnal dipping, and markedly elevated BP variability. It was the strongest independent predictor of non-dipping (adjusted OR = 42.0; Firth OR = 11.73; both p < 0.001) and high SBP variability (adjusted OR = 41.7; Firth OR = 26.29; both p < 0.001). Arterial load substantially improved model discrimination (AUC = 0.819 for non-dipping; 0.979 for SBP variability), whereas adding TyG to arterial load produced minimal incremental benefit. Conclusions: The coexistence of elevated TyG and increased arterial load defines a distinct hemodynamic endotype characterized by severe circadian blood pressure disruption and exaggerated short-term variability. While arterial load emerged as the principal determinant of adverse ambulatory blood pressure phenotypes, TyG alone demonstrated limited discriminative capacity. These findings suggest that TyG primarily acts as a metabolic modifier, amplifying adverse ambulatory blood pressure phenotypes predominantly in the presence of underlying arterial instability rather than serving as an independent discriminator. Integrating metabolic and hemodynamic domains may therefore improve risk stratification and help identify a small but clinically meaningful subgroup of patients with extreme ambulatory blood pressure dysregulation. Full article

Automated recognition of rock mass discontinuities in vegetated high-slope terrains remains a challenging task critical to geohazard assessment and slope stability analysis. This study presents an integrated framework combining close-range UAV photogrammetry with an Improved Superpoint Transformer (ISPT) for semantic segmentation and structural characterization. High-resolution UAV imagery was processed using an SfM–MVS photogrammetric workflow to generate dense point clouds, followed by a three-stage filtering workflow comprising cloth simulation filtering, volumetric density analysis, and VDVI-based vegetation discrimination. Feature augmentation using volumetric density and the Visible-Band Difference Vegetation Index (VDVI), together with connected-component segmentation, enhanced robustness under vegetation occlusion. Validation on four vegetated slopes in Buyun Mountain, China, achieved an overall classification accuracy of 89.5%, exceeding CANUPO (78.2%) and the baseline SPT (85.8%), with a 25-fold improvement in computational efficiency. In total, 4918 structural planes were extracted, and their orientations, dip angles, and trace lengths were automatically derived. The proposed ISPT-based framework provides an efficient and reliable approach for high-precision geotechnical characterization in complex, vegetation-covered rock mass environments. Full article

This paper proposes a robust convolutional neural network (CNN) architecture for human activity recognition (HAR) using smartphone accelerometer data, evaluated on the WISDM dataset. We introduce two novel pooling mechanisms—Pooling A (Extrema Contrast Pooling (ECP)) and Pooling B (Center Minus Variation (CMV))—that enhance feature discrimination and noise robustness. ECP emphasizes sharp signal transitions through a nonlinear penalty based on the squared range between extrema, while CMV Pooling penalizes local variability by subtracting the standard deviation, improving resilience to noise. Input data are normalized to the [0, 1] range to ensure bounded and interpretable pooled outputs. The proposed framework is evaluated in two separate configurations: (1) a 1D CNN applied to raw tri-axial sensor streams with the proposed pooling layers, and (2) a histogram-based image encoding pipeline that transforms segment-level sensor redundancy into RGB representations for a 2D CNN with fully connected layers. Ablation studies show that histogram encoding provides the largest improvement, while the combination of ECP and CMV further enhances classification performance. Across six activity classes, the 2D CNN system achieves up to 96.84% weighted classification accuracy, outperforming baseline models and traditional average pooling. Under Gaussian, salt-and-pepper, and mixed noise conditions, the proposed pooling layers consistently reduce performance degradation, demonstrating improved stability in real-world sensing environments. These results highlight the benefits of redundancy-aware pooling and histogram-based representations for accurate and robust mobile HAR systems. Full article

Introduction: The global prevalence of chronic kidney disease (CKD) is increasing and it is associated with higher mortality rates. Transforming growth factor-beta (TGF-β) can serve as a novel biomarker for early prediction of chronic kidney disease (CKD) progression. Methods: This was a prospective longitudinal study among black patients with CKD who attended the Charlotte Maxeke Johannesburg Academic Hospital between September 2019 and March 2022. Patients provided urine and blood samples for laboratory investigations at study entry (0) and at 24 months follow up. Baseline serum and urine TGF-β1, TGF-β2 and TGF-β3 levels were measured using ELISAs. Multivariable logistic regression analysis was utilized to determine if TGF-β isoforms could predict CKD progression. Results: A total of 312 patients were enrolled at baseline, of whom 275 (88.1%) had early-stage CKD (Stage 1–3). A majority, 95.2% (297/312), of the patients completed the study after 2 years follow up. The prevalence of CKD progression was 47.8% when measured by a sustained decline in eGFR of >4 mL/min/1.73 m²/year or more and 51.9% when measured by a change in uPCR > 30%. The patients with CKD progression had significantly lower eGFR and increased urine protein–creatinine ratios compared to non-progressors. Furthermore, comparing progressors with non-progressors, the median serum TGF-β1 was 21210 (15915–25745) ng/L vs. 24200 (17570–29560) ng/L and the median urine TGF-β3 was 17.5 (5.4–76.2) ng/L vs. 2.8 (1.8–15.3) ng/L, respectively. Baseline serum and urine TGF-β isoforms were unable to discriminate between CKD progressors and non-progressors after multivariable logistic regression analysis. Conclusions: Despite the multiple roles of TGF-β isoforms in kidney disease, baseline levels were not predictive of chronic kidney disease progression. Full article