Search Results (12,158)

The rapid advancements in large language models (LLMs) have transformed natural language processing, enabling their application in diverse domains such as conversational agents and decision-support systems in sensitive areas like healthcare, finance, and eldercare. However, as LLMs are increasingly integrated into real-world contexts, concerns about their adherence to ethical principles, privacy norms, and contextual expectations have become critical. Privacy preservation is particularly pressing in interactions involving personal or sensitive data, where ensuring that LLMs align with societal norms while mitigating risks of information leakage is essential to fostering trust and ensuring responsible deployment. Contextual integrity (CI) provides a robust framework to address these challenges, emphasizing that information flows should adhere to context-specific social norms. This principle is especially vital in sensitive applications, where LLMs must evaluate roles, information attributes, and transmission principles to maintain ethical behavior. Despite their linguistic proficiency, LLMs often fail to recognize and adapt to nuanced contextual norms, a limitation exacerbated by their probabilistic nature and the biases in their training data, which can lead to inappropriate or harmful outputs. Addressing these shortcomings requires rigorous evaluation methodologies and fine-tuning strategies that embed societal and contextual norms into the models. Full article

Unmanned aerial vehicles (UAVs) are increasingly deployed in missions that require high autonomy and reliable decision-making; however, many operational concepts still assume access to GNSS and stable communication with a human operator. In contested environments, this assumption may no longer hold because GNSS degradation, radio-frequency interference, and intentional jamming can disrupt positioning and communication, thereby reducing mission effectiveness and safety. Recent surveys show that operation in GNSS-denied environments remains a major challenge and often requires alternative perception, localization, and control strategies. In response, this article investigates a reinforcement learning (RL)-based decision-support system for the autonomous control of a quadrotor UAV in a three-dimensional simulated environment. Rather than following pre-programmed waypoints, the UAV learns a control policy through interaction with the environment and reward-driven adaptation. The proposed system is designed for mission execution under uncertainty, limited external guidance, and partial observability. Two policy-gradient approaches are implemented and compared: classical REINFORCE and Proximal Policy Optimization (PPO) with an Actor–Critic architecture. The study presents the simulation environment, state and action representation, reward formulation, staged training procedure, and comparative evaluation. The results indicate that, within the considered unseen test scenario, the PPO-based configuration achieved higher mission effectiveness than REINFORCE in the final unseen test scenario, supporting the practical relevance of structured deep reinforcement learning for UAV operation in GPS-denied and communication-constrained environments. Full article

Background/Objectives: Down syndrome (DS), the most common chromosomal disorder, is associated with obesity and related metabolic complications. Although physical activity (PA) improves health outcomes in individuals with DS, global research trends in this field have not been systematically synthesized, and evidence regarding the effects of aerobic training (AT) on obesity-related parameters in individuals with DS remains inconsistent. This study incorporated a dual bibliometric and meta-analytical approach. Methods: First, the bibliometric analysis included 321 original research articles published between 2001 and 2024, retrieved from Scopus, Web of Science, and PubMed. Second, a meta-analysis of 15 randomized controlled trials (n = 477) was conducted to examine the effects of AT on obesity-related parameters, including body weight (BW), body mass index (BMI), fat mass (FM), waist circumference (WC), and waist-to-hip ratio (WHR) in individuals with DS. Results: Keyword co-occurrence and collaboration network analyses revealed a notable increase in research output since 2018, with “adolescent,” “obesity,” and “intellectual disability” the most co-occurring keywords associated with DS and PA. “Obesity” emerged as the most prominently growing keyword associated with DS and PA. A meta-analysis concluded that AT reduced FM (standardized mean differences [SMD] = –0.44; p < 0.001) and WC (SMD = –0.39; p < 0.01), while subtle changes in BW, BMI, and WHR were found. These findings suggest that AT improves body composition, particularly reducing central adiposity, even without changes in traditional weight-based metrics. Conclusions: Our findings demonstrate that AT can be an effective non-pharmacological strategy for improving body composition in individuals with DS and obesity and highlight the urgent need to shift clinical and research paradigms toward multidimensional, individualized health strategies that support PA and healthy body composition throughout the lifespan. Full article

Introduction: Persons diagnosed with Autism Spectrum Disorder (ASD) require adaptations to dental care that many undergraduate programmes may not explicitly treat. This cross-sectional pilot study assessed the extent of ASD-related content in Australia and New Zealand (ANZ) dental and oral health curricula and explored Oral Health Therapy students’ knowledge and self-efficacy. Methods: Online surveys of academic staff across ANZ programmes and Bachelor of Oral Health Therapy students at the University of Newcastle were conducted. Quantitative data was summarised descriptively, and free text responses underwent thematic analysis. Results: Fifteen educator responses (8% of 178 invitees) suggest limited ASD-specific teaching and minimal use of simulation-based education. Among 38 student responses (from one institution), knowledge was generally foundational, but misconceptions persisted and no respondents reported high confidence in providing oral health care for Autistic patients. Interest in further training was high. Conclusions: Within the constraints of low response rates and a single institution student sample, these preliminary findings suggest opportunities to strengthen Autism-related teaching, particularly sensory adaptations, communication strategies, and experiential learning. Inferences should be considered exploratory and hypothesis generating. Limitations: Low educator responses and potential response bias due to limited external validity from a single student cohort. Full article

Reliable, explainable prediction of the maternal mortality ratio (MMR) is challenging in global health because country-level indicators are heterogeneous and missingness is often informative rather than random. This study aims to develop and validate a missingness-aware TabNet (MA-TabNet), an attention-based framework that treats absence patterns as learnable signals while maintaining a stable feature space for country-level MMR forecasting and interpretation. We build a country–year panel from a publicly available global nutrition and health dataset and predict MMR using socioeconomic and health indicators to test whether missingness patterns add predictive signal beyond observed covariates. The model applies a distribution-aware selective masking strategy, adding missingness indicators only for variables with high missing rates; remaining gaps are handled by median imputation, with indicators retained to explicitly encode reporting uncertainty. Country codes and regional groupings are encoded as learnable embeddings, and entmax-based sequential attention is used to improve feature selection via sparse, competition-style masks under correlated determinants. Hyperparameters are tuned using Bayesian optimization, and evaluation follows a temporally realistic protocol (train on earlier years; test on a future held-out year). MA-TabNet achieves a mean absolute error (MAE) of 21.05, root mean squared error (RMSE) of 36.24, and R² of 0.9739, outperforming strong tree-based baselines and improving on the original TabNet while avoiding the training instability observed in some transformer-style tabular models. For transparency, we report attention-derived global and local importance, compare original versus missing-mask features in model importances, and complement these with permutation-based Shapley additive explanation summaries, permutation importance (MAE drop), partial dependence plots for top drivers, and continent-stratified residual analyses to clarify how structural reporting gaps shape predictions and to support trustworthy maternal health monitoring. Overall, these findings suggest that modeling missingness as a measurable reporting signal can yield accurate, auditable forecasts that are better aligned with temporally realistic SDG 3.1 monitoring than “fill-and-forget” preprocessing. Full article

Investigative interviewers play a crucial role in eliciting information from children; therefore, gathering the views and experiences of professionals helps deepen our understanding and guide areas for improvement. Twelve police officers and forensic interviewers, based in the United Kingdom and the United States, were semi-structure interviewed. A thematic analysis was used to analyse the data, leading to the identification of three main themes: challenges and limitations in interview process, strategies for enhancing interview quality and effective techniques for information gathering. Participants noted limited flexibility with respect to minors, technology-related gaps and the significance of third-party disruptions. They called for better training and interview environments with adaptations to fit to children’s variable abilities, sustained rapport and supportive cues. The findings strengthen our understanding of child investigative interviews by providing updated evidence from the professionals who work directly with them. Drawing on the study’s findings, hypotheses were formulated to assess the effectiveness of interview techniques, update guidelines and ultimately improve child protection through more efficient pursuit of justice. Full article

Accurate prediction of mineral concentrations from sparse exploration data is important for resource estimation. This study evaluates hybrid prediction models combining machine learning (ML) and geostatistics to predict aluminum (Al) concentrations. Twelve hybrid configurations were generated by combining six ML backbones—Random Forest, XGBoost, AdaBoost, ResNet, U-Net, and Spatial Transformer Network—with Ordinary Kriging (OK) and Universal Kriging (UK). Model performance was evaluated using 10-fold spatial cross-validation (CV) to reduce spatial leakage, and hyperparameters were tuned by grid-search CV within the training folds. For the hybrid models, residual kriging was fitted using cross-fitted out-of-fold residuals to reduce optimistic bias and prevent information leakage. The results showed no consistent performance separation between OK and UK variants. More importantly, the effect of integration was backbone dependent rather than uniformly beneficial. RF-based predictions showed the strongest overall out-of-sample performance, whereas hybrid gains for other backbones were generally modest. After multiple-comparison correction, most differences between standalone and hybrid models were not statistically significant. These findings indicate that increasing model complexity through hybridization does not guarantee improved accuracy and highlight the importance of spatially explicit, bias-aware evaluation when selecting prediction strategies for mineral resource exploration. Full article

Rapid and reliable tea classification is valuable for routine product screening, yet conventional sensory or physicochemical methods are subjective or time-consuming. Electronic nose (E-nose) sensing provides a fast alternative, but performance often degrades under domain shifts caused by different tea types, commercial categories, or acquisition conditions. This study proposes MGDA-Net, a multi-granularity domain adversarial network for cross-domain tea classification using E-nose time-series signals. MGDA-Net learns local temporal dynamics via a CNN branch and global contextual dependencies via a self-attention branch, and fuses them through an adaptive gating module. A branch-level adversarial alignment strategy is introduced to reduce source–target discrepancy at both local and global feature levels. A three-stage training procedure, consisting of source pretraining, adversarial alignment, and target fine-tuning, enables knowledge transfer from a labeled green tea source-domain to two target tasks. Experiments on oolong tea commercial-category classification (6 classes) and jasmine tea retail price-level classification (8 classes) show that MGDA-Net achieves mean accuracies of 99.31 ± 0.69% and 99.38 ± 0.51% over 10 independent runs, substantially outperforming all compared baseline methods. Ablation studies, feature-space analyses, and label-efficiency experiments further confirm the contribution of each component and show that MGDA-Net maintains mean accuracies above 87% when only 40% of the target-domain labels are used for fine-tuning. These findings suggest that MGDA-Net is a promising approach for cross-domain tea classification using E-nose data. Full article

Road traffic injuries represent one of the most critical public health challenges in the Gulf region. Predicting traffic accident severity is therefore a critical component of evidence-based road safety management. In this study, we develop machine learning frameworks for predicting traffic accident severity using Qatar’s national dataset (2020–2025), addressing extreme class imbalance and interpretability. A dataset of 588,023 accident records was systematically preprocessed from 1,000,500 raw reports. We compare three approaches: multi-class (four severity levels), binary (Safe vs. Severe), and cascaded two-stage (combining both). Six classifiers were evaluated across two encoding methods and three balancing strategies. Systematic hyperparameter tuning with 5-fold stratified cross-validation was performed for all models. The binary LightGBM classifier achieved BA = 71.04%, AUC-ROC = 0.772, Sensitivity = 61.03%, and Specificity = 81.05%, demonstrating superior performance over multi-class approaches. Temporal validation on 2025 data (trained on 2020–2024 data) supported good temporal generalization. Analysis of 10,000 test instances identified the time period as the dominant predictor of accident severity. The binary LightGBM framework provides an interpretable and effective approach for severe accident identification and risk prioritization, with SHAP findings supporting targeted temporal enforcement and pedestrian safety as evidence-based policy priorities. Full article

Reliable change detection in indoor environments remains a challenge for autonomous robotic systems using 3D LiDAR. Existing methods often require manual annotation, computationally intensive architectures, or focus on outdoor scenes. This paper presents SPICD-Net, a lightweight Siamese PointNet framework for indoor 3D change detection trained exclusively on synthetically generated anomalies, eliminating manual labeling. The framework offers three deployment-oriented contributions: a three-class Siamese formulation separating no-change, changed, and geometrically inconsistent tile pairs; a pre-FPS anomaly injection strategy that aligns synthetic training with inference-time preprocessing; and a stochastic-gated Chamfer-statistics branch that complements learned embeddings with explicit geometric cues under consumer-grade hardware constraints. Evaluated on 14 controlled simulation experiments in an indoor corridor dataset, SPICD-Net achieved aggregated Precision = 0.86, Recall = 0.82, F1-score = 0.84, and Accuracy = 0.96, with zero false positives in the no-change baseline and mean inference time of 22.4 s for a 172-tile map on a single consumer GPU. Additional robustness experiments identified registration accuracy as the main operational prerequisite. A limited real-world validation in one unseen room (four scans, 67 tiles) achieved Precision = 0.583, Recall = 1.000, and F1 = 0.737. Full article

Recent geopolitical events have led to an increased research focus on the experiences of female refugees. As careers play a crucial role in socio-economic integration, this study aims to examine the scope and characteristics of research findings on the careers of refugee women in host countries. Following the general research questions for bibliometric analysis, the major trends and intellectual structures of the research field of women refugees’ careers were identified. Four hundred and fifty-three articles selected from the Web of Science database (search by title, abstract, and keywords) for the period 2000–2023 were analyzed using VOSviewer (1.6.20). The results show that key challenges faced by forcibly displaced women include mental health disorders, language barriers, discrimination, downward career mobility, and pressure of traditional gender roles. The research reveals that critical enablers for female refugees’ workforce participation and economic independence are language training, culturally sensitive healthcare, and access to childcare. Simultaneously, empowerment strategies, including entrepreneurship and participation in professional networks, are proved to foster resilience and create pathways for successful career steps. Full article

Neuroimplants represent an emerging medical technology, offering new therapeutic approaches for severe neurological and psychiatric disorders. One of the key limitations to long-term neuroimplant performance is the foreign body response elicited by intracortical implantation. Among the contributing cell types, astrocytes play a central role in glial scar formation around the implant, which can compromise device functionality. Immunofluorescence of glial fibrillary acidic protein (GFAP) provides a well-established marker of astrogliosis (neuroinflammation), yet quantitative and reproducible assessment of astrocyte morphology remains challenging due to the complexity and variability of image analysis approaches. Here, we aimed to quantitatively assess implantation-induced astrogliosis and to determine how classifier training strategy influences segmentation outcomes and morphometric measurements. We present a machine learning-assisted pipeline based on the LabKit plugin in Fiji for segmentation and morphometric analysis of GFAP-positive astrocytes in peri-implant scar versus distant cortical regions. Using this approach, we demonstrate an increase in GFAP expression, cell area, and astrocytic process length as well as the redistribution of GFAP signal along astrocytic processes within scar regions. We show that different classifier training strategies produce systematically distinct segmentation outcomes, with rule-compliant annotation improving agreement with manually defined ground truth. These findings highlight the critical role of annotation strategy in shallow learning-based segmentation and provide a practical framework for improving reproducibility of astrocyte morphometry in studies of neuroinflammation and neuroimplant biocompatibility. Full article

To address the challenge of balancing computational efficiency with fine-grained feature capture in complex field environments when using existing deep learning methods for tomato leaf disease detection, this paper proposes a novel lightweight classification model called Visual Mamba with Frequency-channel attention, Cross-layer attention and Salient feature suppression (VMamba-FCS). Based on the visual state-space model, this model integrates three collaborative enhancement mechanisms: a frequency-domain channel attention module, which improves the perception of disease-related textures by recalibrating features in the frequency domain; a cross-layer attention module, which promotes multi-scale feature fusion by integrating the semantic context of early layers; and a salient feature suppression module, which forces the network to learn more comprehensive discriminative features to improve robustness by suppressing overactivated feature regions during training. Experimental results on the real-world field dataset “Tomato-Village” demonstrate that VMamba-FCS achieves a classification accuracy of 93.62% and an inference speed of 126.5 frames per second (FPS) with only 1.20 M parameters, representing a 7.48% improvement in accuracy compared to the basic VMamba model. In the cross-dataset (PlantDoc) generalization test, VMamba-FCS significantly outperformed all comparison models with an accuracy of 71.3%, demonstrating its excellent domain adaptability and robustness. This work verifies the effectiveness of the multi-mechanism collaborative enhancement strategy in the state-space model architecture, providing a new lightweight solution for real-time and accurate agricultural disease detection on resource-constrained edge devices. Full article

In natural agricultural environments, plant disease monitoring faces significant challenges, including a highly uneven (long-tail) distribution of disease species, tiny scales of early-stage lesions, and complex, variable backgrounds. These factors hinder the ability of existing lightweight models to balance detection accuracy and computational efficiency. To address these issues, this paper proposes a detection scheme driven by the synergy of data distribution reshaping and model architecture optimization. At the data level, we propose the CALM-Aug augmentation strategy. Based on the statistical distribution characteristics of disease categories, this strategy utilizes object-level copy-paste logic to specifically compensate for the feature shortcomings of rare disease samples. It introduces a teacher-guided screening mechanism and employs accept–reject sampling to ensure the pathological consistency of the augmented samples, thereby alleviating the model’s inductive bias toward head categories. At the model architecture level, using YOLOv11 as the baseline, the YOLO11-ARL model adapted to agricultural scenarios is constructed. It enhances sensitivity to early point-like disease spots through Efficient Multi-Scale Convolutional Pyramids and lightweight decoupled detection heads. Furthermore, a Layer-wise Adaptive Feature-guided Distillation Pruning (LAFDP) algorithm is utilized to extract a lightweight version, YOLO11-ARL-PD, achieving a significant reduction in parameters and computational cost. Experimental results on the PlantDoc dataset show that the final model achieves a precision of 89.0% and an mAP@0.5 of 85.3%. Compared to the baseline model YOLOv11n, YOLO11-ARL-PD improves precision and average precision by 7.7 and 2.6 percentage points, respectively, while reducing parameters by 51.93% and weights by 46.15%. Cross-dataset tests prove the good generalization performance of the proposed method. This study indicates that, under lightweight constraints, jointly optimizing the training distribution and model architecture is an effective way to improve plant disease monitoring and to support the edge deployment of smart crop-protection systems. All resources for CALM-Aug are available at wyz-2004/CALM-Aug on GitHub. Full article

Type 1 diabetes (T1D) is an autoimmune disease characterized by T-cell-mediated destruction of pancreatic β-cells. Antigen-specific peptide immunotherapy represents a promising strategy to restore immune tolerance. Reliable identification of relevant T-cell epitopes requires accurate prediction of peptide binding to disease-associated major histocompatibility complex (MHC) molecules. In this study, we developed and validated artificial intelligence (AI)-driven machine learning (ML) predictive models for peptides binding to the NOD mouse-specific MHC class I molecules H-2D^b and H-2K^d and the class II molecule I-A^g7. Balanced datasets of experimentally validated binders and non-binders were compiled, divided into training and test sets, and used to construct position-specific logo models and supervised ML classifiers based on z-scale physicochemical descriptors. External validation demonstrated moderate predictive performance for the logo models (ROC AUC 0.685–0.738), whereas AI models, including Random Forest, Support Vector Machine, and Gradient Boosting, achieved substantially improved discrimination (ROC AUC 0.888–0.906). The validated models were applied to the major T1D autoantigens glutamic acid decarboxylase 65, insulin-1, insulin-2 and zinc transporter 8 and predicted multiple binders, with some overlapping with previously reported immunodominant regions. Selected binders were prioritized for further synthesis and in vivo immunogenicity testing in NOD mice. Full article