Search Results (4,913)

Compact low-pressure emitters (LPEs) can improve the affordability of drip irrigation, but they must also demonstrate clog resistance for long-term reliability and adoption. Recent research on LPEs has focused on their hydraulic modeling and characterization, but few studies have evaluated or improved their clog resistance. To address this gap, we present a design theory for clog-resistant LPEs and characterize their performance in the lab and field. We focused on the emitters’ weir (or ‘overflow groove’ or ‘channel’), a micrometer-scale internal hydraulic passage, traditionally having a rectangular cross-section. In LPEs, the weir must be shallow to generate the hydraulic resistance required for low-pressure operation, thereby increasing the risk of particulate-jamming-based clogging. A hydraulic model of weirs with arbitrary cross-sections was used to estimate that trapezoidal profiles could be 33–41% deeper than hydraulically equivalent rectangular ones, suggesting that the trade-off between clog resistance and hydraulic performance in LPEs could be navigated through weir cross-section design. To practically validate this proposition, two compact LPEs with trapezoidal weirs (1 and 2 L/h nominal discharge) were designed and tested in the lab and field. Lab results indicated compatibility with 125 $\mathsf{\mu }$m (1 L/h) and 177 $\mathsf{\mu }$m (2 L/h) mesh filters that are typical for these flow rates, providing a basis for field testing the LPEs against commercial emitters. After field tests with these filters, the LPEs held 90–94% of their initial discharge and demonstrated irrigation reliability that was statistically on par with or better than some commercial emitters, despite having 15–65% lower operating pressure. The findings of this work demonstrate the practical viability of compact LPEs for affordable drip irrigation and provide a design framework for their continued development. Full article

This study evaluates the potential of L-band passive microwave data for monitoring soil moisture (SM) in boreal and temperate forests using SMAP and SMOS AM and PM overpasses. SMAP and SMOS Level 3 SM products were first assessed for spring and summer seasons. SMOS showed lower accuracy (r² = 0.04–0.24, ubRMSE = 0.09–0.13 m³/m³), while SMAP performed better (r² = 0.18–0.62, ubRMSE = 0.05–0.07 m³/m³) across sites and overpasses. Given the larger number of SMAP TB observations at a fixed incidence angle and greater temporal coverage over the study area, SMAP was selected for SM estimation using ML models. Feature importance analysis identified brightness temperature (TB) as the most influential variable, followed by vegetation water content (VWC), air and soil temperatures, and the microwave polarization difference index (MPDI). Soil and air temperatures were interchangeable during AM overpasses, whereas PM overpasses showed distinct differences, likely due to thermal absorption by dense vegetation. Using optimal features, SM was estimated with CatBoost, Gradient Boosting (GB), Random Forest (RF), and Principal Component Regression (PCR), using stratified shuffle split (SSS) and leave-one-year-out cross-validation (LOYOCV). In SSS, CatBoost achieved slightly higher accuracy than the other ensemble models (AM: r² = 0.73; PM: R² = 0.74), while PCR yielded substantially lower accuracy across both overpasses. LOYOCV showed closer rankings among models, with CatBoost ranking highest overall (r² = 0.58 for AM and 0.54 for PM). Results highlight the feasibility of improved SM estimation in forests using L-band TB, VWC, temperature variables, and MPDI. Full article

Underwater RGB–sonar object detection remains challenging due to severe optical degradation, strong sonar noise, and spatial misalignment between heterogeneous modalities. Existing multimodal detectors usually rely on simple feature aggregation or limited structural coupling, which cannot effectively model global cross-modal dependencies or address modality-specific degradation. To address these challenges, we propose Chrs-Net, a YOLOv12-based dual-stream framework for underwater RGB–sonar object detection. The proposed network integrates three key components: a Transformer-based Cross-Modal Communication Fusion module (C-mcf) for global cross-modal interaction and semantic alignment, a Multi-Layer Feature Enhancement module (MLFE) for degraded optical feature enhancement, and a Pinwheel-Shaped Convolution module (PConv) for sonar-side structural feature extraction. In addition, an RGB–sonar object detection dataset is constructed for experimental evaluation by relabeling part of the RGBS benchmark, combining simulator-collected samples, and introducing style-transfer-based augmentation to improve data diversity. Experiments on the constructed dataset yield 94.91% mAP@0.5 and 61.10% mAP@0.5:0.95 on the RGB branch, and 94.00% and 57.13% on the sonar branch, respectively, with an inference speed of 53.6 FPS. Compared with representative single-modality and multimodal detectors, Chrs-Net consistently yields superior detection accuracy and localization performance. These results demonstrate that the combination of global cross-modal communication and modality-specific enhancement is effective for robust underwater RGB–sonar object detection in complex environments. Full article

Background: Age at diagnosis and histologic differentiation are clinically relevant in early gastric cancer (GC), as poorly differentiated tumors and those diagnosed in younger patients often demonstrate more aggressive characteristics. Serum microRNAs (miRNAs) may provide insights into the molecular basis of these features. Methods: We compared expression profiles between undifferentiated and differentiated early GC cases to identify differentially expressed miRNAs (DEmiRNAs) and associated enriched pathways. Using Lasso regression, we developed and cross-validated a histologic differentiation classifier based on miRNA profiles from 1399 early GC serum samples. Finally, cancer-specific miRNA differences between adolescent and young adult (AYA) and non-AYA patients were evaluated using samples from cancer cases and normal controls. Results: We identified 75 differentiation-associated DEmiRNAs targeting genes enriched in cancer hallmark pathways such as TP53 and PI3K/AKT/mTOR signaling. In the validation set, the combined Lasso model predicted differentiation status with a sensitivity of 69.2%, specificity of 75.3%, positive predictive value (PPV) of 66.9%, negative predictive value (NPV) of 77.2%, an overall accuracy of 73.1%, and an area under the curve (AUC) of 79.7%. Comparison of AYA and non-AYA groups identified 52 cancer-specific and age-related miRNAs. Notably, three components of a previously reported four-miRNA GC diagnostic signature were significantly associated with age. Conclusions: Age-related variation in miRNA expression suggests that patient age may influence the performance of the existing four-miRNA diagnostic signature in early GC. Overall, our findings demonstrate the utility of miRNA profiling for predicting differentiation status in early GC and reveal age-associated variation in cancer-specific miRNAs. Full article

Loquat (Eriobotrya japonica) is a commercially important subtropical fruit, and its internal sweetness is an important indicator of market quality. Accurate and non-destructive determination of total soluble solids content (TSSC) is therefore essential for fruit grading and quality evaluation. In this study, short-wave infrared hyperspectral imaging (1000–2400 nm) was combined with a multi-scale spectral attention adaptive convolutional neural network (MSSA-ACNN) for rapid TSSC prediction. Spectral data were preprocessed using an SG-MSC-DT strategy to reduce noise and scattering effects, while conventional models (PLSR, Ridge, and SVM) were used for comparison. The proposed model combines multi-scale feature extraction with a dual-path attention mechanism, enabling adaptive enhancement of informative chemical wavebands while suppressing irrelevant variations. Experimental results, rigorously validated through a 5-fold cross-validation strategy, demonstrated that the proposed approach achieved the best predictive performance, with an R_p² of 0.942, RMSEP of 0.505, and RPD of 3.091, outperforming traditional methods. In addition, attention weight analysis revealed that the model mainly focused on spectral regions associated with water and carbohydrate absorption, indicating consistency between the learned features and known chemical information. These results suggest that the proposed method provides an effective and interpretable approach for non-destructive evaluation of loquat quality and shows potential for application in intelligent fruit grading systems. Full article

Background/Objectives: Evidence for herbal ergogenic aids remains uncertain, and ashwagandha trials span heterogeneous performance domains. This review evaluated oral Withania somnifera supplementation on exercise performance and explored participant-, outcome-, formulation-, and supplementation-related moderators. Methods: PubMed, Web of Science, Cochrane Library, Embase, and SPORTDiscus-EBSCO were searched from inception to 1 April 2026. Eligible randomized controlled trials compared oral ashwagandha with placebo or control conditions and reported objective exercise-performance outcomes. Dependent effects were synthesized using restricted-maximum-likelihood three-level random-effects models; 95% prediction intervals, GRADE certainty ratings, subgroup analyses, and dose/duration meta-regressions were reported. Results: Thirteen trials involving 599 participants contributed 79 effect sizes. Samples were mainly young adults or athletes; reported ages included one 18–40-year trial and one late-adolescent athlete cohort aged 17.4 ± 1.7 years. Trial-level sex composition was four male-only, one female-only, three mixed-sex, and five incompletely reported cohorts. Ashwagandha improved overall exercise performance on average (Hedges’ g = 0.47, 95% CI [0.25, 0.69], p < 0.001; I² = 60%; 95% prediction interval [−0.40, 1.33]), but the prediction interval crossed zero. Exercise type was the clearest moderator (P_between = 0.006): evidence was most consistent for aerobic endurance (g = 0.54, 95% CI [0.22, 0.85], p = 0.002), whereas strength effects were positive but uncertain and power or muscular endurance evidence remained sparse. Dose analyses were hypothesis-generating; 500–600 mg/day was the most evidence-supported extract-dose range. Conclusions: Oral ashwagandha may improve selected exercise-performance outcomes, particularly aerobic endurance, but benefits are not uniform across contexts. Future trials should be preregistered, adequately powered, double-blind, formulation-standardized, sex-stratified, and include rigorous blinding checks, mechanistic endpoints, adverse-event monitoring, and sport-specific performance tests. Full article

Multimodal biometric systems have become essential components of modern electronic identity and authentication platforms where robustness under real-world degradation is critical. However, opaque face masks impose severe facial occlusion and attenuate high-frequency spectral components. Conversely, transparent face shields introduce complex specular reflections and act as an acoustic channel distortion source. Addressing these asymmetric degradation challenges, this paper proposes a reliability-aware Dynamic Score Fusion (DSF) for multimodal biometric identification. The proposed method performs sample-level reliability estimation for both face and voice modalities at the input stage. This enables sample-wise adaptive weighting of modality scores based on their estimated reliability. The framework integrates an ElasticFace-Arc backbone for face recognition with an Emphasized Channel Attention, Propagation and Aggregation—Time Delay Neural Network (ECAPA-TDNN) for speaker identification. The proposed approach is evaluated on the FaciaVox dataset, comprising face images and voice recordings acquired under multiple face-covering conditions. Experiments under the Standard to Cross-Condition Protocol (SCCP) and Multi-Condition Protocol (MCP) demonstrate that the proposed DSF consistently outperforms conventional score-level fusion methods, including Weighted Sum Fusion (WSF) and Logistic Regression Fusion (LRF). It achieves average Rank-1 accuracies of 89.6% (SCCP) and 93.7% (MCP), with gains of up to 9.3 percentage points over these baselines. The reliability estimators further demonstrate strong predictive capability, yielding Area Under the Curve (AUC) values above 0.95 for both modalities in distinguishing correctly and incorrectly identified samples under the closed-set identification setting. These findings confirm that sample-wise reliability modeling provides an effective mechanism for enhancing multimodal biometric performance under challenging mask and shield conditions, supporting the deployment of robust AI-driven electronic identification systems. Full article

Monitoring water surface algal blooms from surveillance perspectives faces challenges such as small objects, low texture contrasts, dynamic background interferences, and limited labeled datasets. In this study, we propose GECA-YOLOv26, a lightweight model that integrates Ghost Convolution (GhostConv) and Efficient Channel Attention (ECA) modules. First, the GhostConv lightweight module is introduced in the first layer of the YOLOv26 backbone, reducing parameters from 4608 to 2704 and achieving a 41% reduction in computational cost. Second, eight ECA modules are embedded at key locations after backbone downsampling and neck feature fusion to enhance feature representation and mitigate degradation caused by model lightweighting. Finally, the MuSGD optimizer is used for training, with adaptive modifications to resolve tensor shape conflicts with the ECA modules. Experimental results indicate that the model achieves a mAP50 of 82.16%. Compared with the YOLOv26 baseline, our model improves mAP50 by 6.42%, while mAP@0.5:0.95 decreases by 0.79% and inference speed reduces from 143 FPS to 123 FPS. The model also reduces parameters and size, achieving 5.19 MB and 1864 fewer parameters. Compared with YOLOv8, YOLOv10, and YOLOv11, the proposed model improves mAP50 by 2.12%, 5.99%, and 2.79%, respectively. To evaluate the stability of the results under small-sample conditions, we conducted 3-fold and 5-fold cross-validation experiments, which demonstrated that the model performs robustly across different folds and random seeds. Ablation studies further confirm the effectiveness of each module. Heatmap analysis demonstrates that the proposed model effectively highlights small object regions, remains robust under limited-sample conditions, and reduces model complexity. This study provides a novel solution for algal bloom detection in surveillance scenarios. Full article

High-resolution spatiotemporal sea ice concentration (SIC) estimates are essential for Arctic navigation and ice analysis, but existing observational products are often too coarse, and physics-based models are computationally expensive. This study proposes a data-driven framework that couples Light Gradient Boosting Machine (LightGBM) temporal prediction with Kriging-based spatial interpolation to reconstruct SIC fields over the Northern Sea Route sector. LightGBM is trained on a grid-based SIC time series with engineered features representing persistence, seasonality, and short-term variability, enabling multi-horizon forecasting across large spatial grids. The predicted SIC fields are then refined using Ordinary Kriging (OK) and Co-Kriging (CK) with Gaussian and spherical semi-variogram models. Prediction performance is evaluated using root mean square error, and interpolation accuracy is assessed through cross-validation. Results show that, for high-latitude regions and resolutions finer than 0.25° × 0.25°, OK with a spherical semi-variogram achieves lower interpolation errors than CK and Gaussian-based alternatives. By sequentially coupling temporal learning and spatial refinement, the proposed framework improves temporal continuity, spatial structure, and error quantification, providing high-resolution SIC information suitable for large-scale Arctic ice analysis and navigation support. Full article

The ski-jump spillway is an energy-dissipating structure that discharges extra water beyond the dam’s capacity. The scour process occurs below spillways due to the collision of the water jet with high energy. It is critical to acquire information on scour holes to improve the dam’s safety and related components. Machine learning (ML) techniques have successfully demonstrated their effectiveness for modeling scour in hydraulic engineering. The present research considers novel approaches of ML models for estimating the scour hole geometries below ski-jump bucket spillways. This study investigates the capability of two novel feature-engineering approaches, namely Stronger Variable Creator Machine (SVCM) and High Correlated Variables Creator Machine (HCVCM), along with Gene Expression Programming (GEP) and their hybrid forms (SVCM+GEP and HCVCM+GEP), which were employed to predict normalized scour depth, scour length, and scour width below ski-jump spillways. Statistical metrics, graphical analyses, the Rank Mean (RM) method, the cross-validation approach, and $U_{95}$ index were used for the evaluation and reliability assessment of the proposed ML models. The results showed that hybrid ML models consistently outperformed individual algorithms. The results indicated that the SVCM+GEP method with $\mathrm{R}\mathrm{M}=1.83$ and $1.50$ had the highest performance compared to other methods for the prediction of ${\displaystyle \frac{Ds}{Dw}}$ and ${\displaystyle \frac{L_s}{D_w}}$, respectively. In addition, the HCVCM+GEP method with $\mathrm{R}\mathrm{M}=1.33$ was the best model for the prediction of ${\displaystyle \frac{Ws}{Dw}}$. In comparison with the conventional regression-based equations and previously reported ML methods, the proposed hybrid approaches improved the prediction results. In addition, the cross-validation method confirmed the robustness and generalization capability of the suggested hybrid ML models. The superior performance of the hybrid models is attributed to their ability to capture complex nonlinear interactions among hydraulic and geometric variables. The developed SVCM/HCVCM+GEP models provide accurate approaches for predicting scour parameters in hydraulic structures. Full article

Foundation models (FMs) are increasingly proposed as general-purpose solutions for computational pathology, with the potential to simplify clinical artificial intelligence deployment by reducing the need for task-specific architectures. However, their reliability across cancer domains with distinct morphological characteristics remains unclear, limiting confidence in real-world clinical use. We benchmarked seven general-purpose pathology FMs and three domain-specific FMs across eleven patch-level datasets spanning three clinically relevant domains: pediatric hematology, prostate cancer, and breast cancer, using both linear probing and last-layer fine-tuning adaptation strategies. By jointly evaluating pediatric leukemia, male-predominant prostate cancer, and female-predominant breast cancer, this study is, to our knowledge, the first to explicitly examine specialist-versus-generalist FM behavior across age- and sex-stratified cancer populations. Performance differences were strongly domain dependent. In hematology, the specialist FM DINOBloom matched and, in several datasets, marginally exceeded leading generalist models (AUC 0.990–0.999 vs. GigaPath 0.981–1.000), suggesting advantages for highly distinctive cellular morphology. In prostate cancer grading, the generalist FM UNI2-h consistently outperformed the specialist HistoEncoder (AUC 0.956–0.977 vs. 0.908–0.964). In breast cancer, UNI2-h achieved the best overall performance across all tasks. No publicly available breast-cancer-specific FM currently exists for direct comparison; therefore, breast cancer results characterize general FM transferability rather than specialist-versus-generalist differences. Importantly, cross-dataset experiments revealed substantial performance degradation under dataset shift in both prostate and breast cancer, indicating that current FMs are not yet robust enough for heterogeneous multi-site clinical use. These findings support the use of generalist FMs as efficient backbones for well-characterized single-site, patch-level tasks, while challenging the assumption that high benchmark performance necessarily reflects true clinical readiness and demonstrating that pathology FMs are not uniformly superior to specialist models. Full article

Background/Objectives: Post-Endoscopic Retrograde Cholangiopancreatography Pancreatitis (PEP) is a clinically significant complication of ERCP, occurring in approximately 2–10% of general cases and at higher rates in high-risk patients. Early prediction of PEP risk may support timely intervention and improved patient management. This retrospective single-center study comparatively evaluated tabular clinical data, endoscopic image data, and multimodal fusion approaches for PEP prediction. Methods: Retrospective data collected from the Sindh Institute of Advanced Endoscopy and Gastroenterology were analyzed using machine learning and deep learning techniques. XGBoost(version 3.2.0) was applied to tabular clinical data, while EfficientNet-B0, ResNet50, and DenseNet201 were used for endoscopic image analysis. A multimodal contrastive learning (MMCL)-based framework combining ResNet50 image features with multilayer perceptron (MLP)-based tabular features was additionally implemented for binary PEP prediction. Class imbalance mitigation techniques, including data augmentation and balancing strategies, were applied during training. Model performance was evaluated using the area under the receiver operating characteristic (ROC) curve (AUC), sensitivity, F1-score, and precision. SHAP analysis was performed to identify important predictive features. Results: The tabular XGBoost model achieved the best predictive performance with an AUC of 0.95 and a sensitivity of 0.50, while five-fold cross-validation yielded an AUC of 0.79 and a sensitivity of 0.48. Among image-based models, ResNet50 achieved the highest performance, with an AUC of 0.76 and a sensitivity of 0.40. The multimodal model achieved an AUC of 0.57 and a sensitivity of 0.20. SHAP analysis identified cannulation time, ampulla type, and age as prominent features associated with PEP prediction. Conclusions: This exploratory study suggests that structured clinical data currently provide stronger predictive signals for PEP prediction than the available image and multimodal data within this limited cohort. The relatively low occurrence of PEP contributed to class imbalance despite mitigation strategies. Future multicenter studies with larger datasets, improved image availability, synthetic data generation, and advanced multimodal fusion techniques may improve predictive performance and clinical applicability. Full article

Thematic analysis of large-scale political discourse remains a challenge due to semantic complexity and overlapping policy areas and changing diplomatic vocabulary. Although large language models (LLMs) offer promise for scalable thematic classification, their reliability in politically sensitive contexts requires systematic validation against expert human annotations. We evaluate LLM-based thematic classification of United Nations General Assembly (UNGA) speeches across a decade (2014–2023), using 7680 human-annotated themes mapped into 12 policy domains. Our results show that DeepSeek R1 achieves the highest accuracy 77% (F1 = 0.73), followed by ChatGPT, Gemini and LLaMA, with strong performance in lexically stable domains but substantial degradation in semantically overlapping categories such as governance and international cooperation. A unique dimension of our work is timeline analysis, which shows that the performance of LLMs over the years varies strongly and the precision decreases during times of rhetorical transformation, including pandemic-related discussions and the discourses of cooperation determined by the Russia–Ukraine conflict. By linking domain-level ambiguity and geopolitical shifts to temporal instability, this study introduces a dynamic robustness perspective for evaluating LLMs in computational political discourse analysis. Full article

Wearable inertial sensing has considerable potential for process-level analysis of upper-limb function, but further evidence is needed to understand how it can be applied within ecologically structured immersive virtual reality (VR) tasks. Most VR-based functional assessments rely primarily on outcome-level indicators, such as task completion time, success rate, or error count, which may not fully capture how a task is executed. This exploratory study investigated whether wearable IMU signals collected during an immersive VR sushi-making task could support binary detection of a core upper-limb manipulation phase and provide additional information about task execution beyond global performance outcomes. A total of 45 participants contributed usable motion recordings for this study, with five Xsens DOT sensors placed on the hands, forearms, and waist. Three signal modalities were analysed, including acceleration (ACC), gyroscope angular velocity (GYR), and Euler angles. The downstream recognition problem was formulated as a binary classification task (Placing vs. Non-Placing), and a self-supervised learning (SSL) pretrain–fine-tune strategy was evaluated against conventional machine learning and from-scratch deep learning baselines using five subject-wise validation splits. The strongest overall performance was achieved with hand-mounted accelerometer signals, with LeftHand–ACC achieving a Macro-F1 of $0.712\pm 0.128$ and RightHand–ACC achieving $0.679\pm 0.118$. Under both hand-ACC settings, SSL fine-tuning showed higher mean Macro-F1 than the Balanced Random Forest baseline and the same deep architecture trained from scratch. Recognition performance varied substantially across sensor locations, signal modalities, and task segments, with distal upper-limb sensors generally outperforming waist-based configurations. Cross-age analyses further showed that within-cohort and cross-cohort performance did not fully align, indicating sensitivity to age-related distribution shift. Beyond classification, Log Dimensionless Jerk (LDLJ) derived from the Placing action showed a significant positive association with Cognitron motor control time cost ($r=0.636$, $p<0.001$). These findings suggest that wearable IMU sensing can provide preliminary process-level information during immersive VR functional tasks, including task-phase detection, sensing-configuration comparison, cross-cohort generalisation assessment, and exploratory motion-quality analysis. The results should be interpreted as evidence of feasibility rather than as a mature biomechanical or clinical assessment model. Full article

Accurate automatic classification of seed varieties is important for seed sorting, quality assurance, and plant breeding, yet reliable discrimination remains difficult when cultivars exhibit highly similar visual characteristics. This study presents a reproducible and interpretable framework for the binary classification of two Turkish pumpkin seed varieties using tabular morphological descriptors extracted from segmented seed images. Unlike many previous machine learning studies in this domain, which offer limited interpretability and leave model decisions largely as a black box, the proposed approach places Explainable Artificial Intelligence (XAI) at the center of the analysis. The framework combines biologically meaningful feature engineering, Optuna-based hyperparameter optimization, repeated stratified cross-validation, and a comparative evaluation of XGBoost, LightGBM, and CatBoost. Model explainability was investigated using SHapley Additive exPlanations (SHAP) to identify the morphological traits driving both global and instance-level predictions, while corrected repeated k-fold t-tests were used to assess the statistical significance of performance differences, which confirmed comparable accuracy among the three boosting models and a significant advantage over the baseline classifiers. All three boosting ensembles consistently outperformed the baseline classifiers (SVM, Logistic Regression, and Random Forest) on the hold-out test set. CatBoost achieved the best overall results, with an accuracy of 0.888, an F1-score of 0.879, and an MCC of 0.777. SHAP analysis consistently highlighted compactness, roundness, eccentricity, and engineered interaction descriptors as the most influential predictors. Overall, the proposed XAI-driven framework provides an accurate and transparent solution for pumpkin seed classification. Full article