Search Results (1,949)

Skilled operation of forestry harvesters is essential for ensuring safety, efficiency, and sustainability in logging practices. However, conventional training methods are often prohibitively expensive and limited by access to specialized equipment. This study delivers one of the first user-centered validations of a low-cost, VR HMD-based forestry harvester simulator, directly addressing access and scalability barriers in training. With 26 participants, we quantify cognitive load, usability, user experience, and simulator sickness using established instruments. An increase in cognitive load was seen from baseline tutorial to each training module (NASA-TLX: $18.65\to 34.26\to 38.43$; rm-ANOVA, p < 0.001). Usability was ‘Good’ (with a mean SUS score: 76.63), hedonic UX ranked in the top decile (UEQ-S), and simulator sickness was moderate (mean SSQ score: 28.91), while task success remained high across all modules. These results indicate early-stage feasibility and usability of a low-cost VR HMD harvester simulator for student-focused introductory instruction, and they provide actionable design guidance (e.g., managing extraneous load, comfort safeguards) advancing evidence-based VR HMD-based training in the forest engineering and harvesting domain. Our findings validate the potential of VR-HMD as a tool for forestry education capable of addressing training accessibility gaps and enhancing learner motivation through immersive experiential learning. Full article

Lanthanide-based single-molecule magnets are promising candidates for potential applications. Their magnetism is governed by ligand-field splittings, which may require up to 27 ligand-field parameters for accurate modeling. Determining these parameters reliably from measured data is a major challenge, for which machine learning approaches offer promising solutions. We provide an overview of these approaches and present our perspective on addressing the inverse problem relating experimental data to ligand-field parameters. Previously, a machine learning architecture combining a variational autoencoder (VAE) and an invertible neural network (INN) showed promise for analyzing temperature-dependent magnetic susceptibility data. In this work, the VAE-INN model is extended through data augmentation to enhance its tolerance to common experimental inaccuracies. Focusing on second-order ligand-field parameters, diamagnetic and molar-mass errors are incorporated by augmenting the training dataset with experimentally motivated error distributions. Tests on simulated experimental susceptibility curves demonstrate substantially improved prediction accuracy and robustness when the distributions correspond to realistic error ranges. When applied to the experimental susceptibility curve of the complex ${\mathrm{Al}}_2^{\mathrm{III}}$${\mathrm{Er}}_2^{\mathrm{III}}$, the augmented VAE–INN recovers ligand-field solutions consistent with least-squares benchmarks. The proposed data augmentation thus overcomes a key limitation, bringing the ML approach closer to practical use for higher-order ligand-field parameters. Full article

Background: Exercise-induced fatigue (EIF) impairs performance and recovery and may contribute to overreaching/overtraining and adverse health outcomes. Beyond classical explanations (substrate depletion, metabolite accumulation, oxidative stress), accumulating evidence indicates that the gut microbiota modulates fatigue-related physiology through metabolic, immune, barrier, and neurobehavioral pathways. Methods: We conducted a structured narrative review of PubMed and Web of Science covering 1 January 2015 to 30 November 2025 using predefined keywords related to EIF, gut microbiota, recovery, and nutritional interventions. Human studies, animal experiments, and mechanistic preclinical work (in vivo/in vitro) were included when they linked exercise load, microbial features (taxa/functions/metabolites), and fatigue-relevant outcomes. Results: Across models, high-intensity or prolonged exercise is consistently associated with disrupted gut homeostasis, including altered community structure, reduced abundance of beneficial taxa, increased intestinal permeability, and shifts in microbial metabolites (e.g., short-chain fatty acids). Evidence converges on four interconnected microbiota-mediated pathways relevant to EIF: (1) energy availability and metabolic by-product clearance; (2) redox balance and inflammation; (3) intestinal barrier integrity and endotoxemia risk; and (4) central fatigue and exercise motivation via microbiota–gut–brain signaling. Nutritional strategies—particularly targeted probiotics, prebiotics/plant polysaccharides, and selected bioactive compounds—show potential to improve fatigue biomarkers and endurance-related outcomes, although effects appear context-dependent (exercise modality, baseline fitness, diet, and baseline microbiota). Conclusions: Current evidence supports a mechanistic role of the gut microbiota in EIF and highlights microbiota-targeted nutrition as a promising adjunct for recovery optimization. Future work should prioritize causal validation (e.g., fecal microbiota transplantation and metabolite supplementation), athlete-focused randomized trials with standardized fatigue endpoints, and precision approaches that stratify individuals by baseline microbiome features and training load. Full article

Domain incremental learning for semantic segmentation has gained lots of attention due to its importance for many fields including urban planning and autonomous driving. The catastrophic forgetting problem caused by domain shift has been alleviated by structure expansion of the model or data rehearsal. However, these methods ignore similar contextual knowledge between the new and the old data domain and assume that new knowledge and old knowledge are completely mutually exclusive, which cause the model to be trained in a suboptimal direction. Motivated by the prompt learning, we proposed a new domain incremental learning framework named RS-VDP. The key innovation of RS-VDP is to utilize a visual domain prompt to change the optimization direction from input data space and feature space. First, we designed a domain prompt based on a dynamic location module, which applied a visual domain prompt according to a local entropy map to update the distribution of the input images. Second, in order to filter the feature vectors with high confidence, a representation feature alignment based on an entropy map module is proposed. This module ensures the accuracy and stability of the feature vectors involved in the regularization loss, alleviating the problem of semantic drift. Finally, we introduced a new evaluation metric to measure the overall performance of the incremental learning models, solving the problem that the traditional evaluation metric is affected by the single-task accuracy. Comprehensive experiments demonstrated the effectiveness of the proposed method by significantly reducing the degree of catastrophic forgetting. Full article

Wire-directed energy deposition (W-DED) produces steep thermal gradients and rapid heating-cooling cycles due to the moving heat source, where modest variations in process parameters significantly alter heat input per unit length and therefore the full thermal history. This sensitivity makes process tuning by trial-and-error or repeated FE sweeps expensive, motivating inverse analysis. This work proposes an inverse thermal process design framework that couples single-track experiments, a calibrated finite element (FE) thermal model, and a parametric physics-informed neural network (PINN) surrogate. By using experimentally calibrated heat-loss physics to define the training constraints, the PINN learns a parameterized thermal response from physics alone (no temperature data in the PINN loss), enabling inverse design without repeated FE runs. Thermocouple measurements are used to calibrate the convection film coefficient and emissivity in the FE model, and those parameters are used to train a parametric PINN over continuous ranges of arc power (1.5–3.0 kW) and travel speed (0.005–0.015 m/s) without using temperature data in the loss function. The trained PINN model was validated against the calibrated FE model at 3 probe locations with different power and travel speed combinations. Across these benchmark conditions, the mean absolute errors are between 6.5–17.4 °C, with cooling-tail errors ranging from 1.8–12.1 °C. The trained surrogate is then embedded in a sampling-based inverse optimization loop to identify power-speed combinations that achieve prescribed interlayer temperatures at a fixed dwell time. For target interlayer temperatures of 100, 130, and 160 °C with a 10 s dwell time, the optimized solutions remain within 3.3–5.6 °C of the target according to the PINN, while FE verification is within 4.0–6.6 °C. The results demonstrate that a physics-only parametric PINN surrogate enables inverse thermal process design without repeated FE runs while establishing a single-track baseline for extension to multi-track and multi-layer builds. Full article

Sustainable service excellence in globalized industries requires organizations to develop workforce capabilities that support long-term relationship-building, cultural respect, and effective cross-cultural communication. This study examines how cultural intelligence functions as a mechanism for sustainable cross-cultural workforce development by investigating relationships among individual cultural values, cultural intelligence dimensions, and influence tactics among airline cabin crew members. Integrating Hofstede’s cultural dimensions framework, Ang and colleagues’ cultural intelligence model, and Yukl’s influence tactics taxonomy, we test a comprehensive mediation model using survey data from six hundred and sixty-three cabin crew members employed by international airlines operating in Turkey. The findings reveal that collectivism, long-term orientation, and uncertainty avoidance positively predict cultural intelligence development, creating foundations for sustainable cross-cultural competence. Cultural intelligence dimensions demonstrate differentiated effects on influence tactics, with metacognitive and behavioral cultural intelligence enhancing rational persuasion, behavioral cultural intelligence exclusively predicting relational tactics, and complex competitive mediation patterns for coercive tactics wherein motivational cultural intelligence reduces pressure-based influence while cognitive and behavioral dimensions increase strategic assertiveness. Cultural values directly influence tactics beyond cultural intelligence effects, with uncertainty avoidance most strongly predicting both rational and relational approaches that support relationship sustainability, while masculinity and power distance drive coercive tactics that may undermine long-term service relationships. These findings demonstrate that cultural intelligence functions as a multidimensional mediating mechanism with sometimes opposing effects, challenging assumptions that cross-cultural competencies uniformly produce sustainable outcomes. The research contributes to sustainable human resource management theory by illuminating how cultural socialization influences behavioral outcomes through complex psychological pathways, while offering practical guidance for aviation industry recruitment, training, and performance management systems seeking to build sustainable cross-cultural service capabilities. By revealing that certain cultural intelligence dimensions can enable both relationship-building and strategic coercion, the study highlights the importance of coupling cross-cultural skill development with ethical frameworks and motivational engagement to ensure that enhanced cultural capabilities support rather than undermine sustainable, respectful cross-cultural service relationships. Full article

Background/Objectives: Valve replacement (VR) patients are at high risk of postoperative complications, but reproducible Electronic Health Record (EHR) benchmarks for evaluating sequential AI models in this setting are lacking. We develop a reproducible pipeline that extracts two EHR datasets from MIMIC-IV (a general-purpose and a predictive benchmark dataset) capturing perioperative histories, high-resolution time-series, and clinically motivated outcome labels. Methods: The cohort comprises 3890 VR patients with clinician-guided feature selection across diagnoses, procedures, laboratory measurements, medications, and physiological monitoring. As an exemplary use case, we define ICU readmission at first ICU discharge as a surrogate for postoperative risk and derive a predictive benchmark under strict label-leakage control. We then compare a Transformer model trained on tokenized longitudinal EHR sequences with Transformer and XGBoost baselines trained on aggregated feature statistics, and assess performance differences using paired statistical tests across validation splits. Results: ICU readmission stratified in-hospital and 100-day outcomes, including mortality, complications, and rehospitalization, confirming the clinical relevance of the prediction target. The sequential Transformer achieved 0.87 AUROC and 0.69 AUPRC. Corrected resampled t-tests confirm improved performance over the non-sequential Transformer, while the comparison with XGBoost indicates a favorable trend without conclusive evidence. Conclusions: Our findings suggest that leveraging longitudinal EHR sequences yields higher predictive performance than static feature summaries for postoperative risk prediction. The publicly released preprocessing pipeline and cohort-construction code enable researchers with MIMIC-IV access to reproduce the datasets and provide a robust benchmark for developing and comparing time-series models in post-valve replacement care. Full article

Background/Objectives: Accurate and reliable automated dermoscopic lesion classification remains challenging. This is due to pronounced dataset bias, limited expert-annotated data, and poor cross-dataset generalization of conventional supervised deep learning models. In clinical dermatology, these limitations restrict the deployment of data-driven diagnostic systems across diverse acquisition settings and patient populations. Methods: Motivated by these challenges, this study proposes a transformer-based, dermatology-specific foundation model. The model learns transferable visual representations from large collections of unlabeled dermoscopic images via self-supervised pretraining. It integrates large-scale dermatology-oriented self-supervised learning with a hierarchical vision transformer backbone. This enables effective capture of both fine-grained lesion textures and global morphological patterns. The evaluation is conducted across three publicly available dermoscopic datasets: ISIC 2018, HAM10000, and PH2. The study assesses in-dataset, cross-dataset, limited-label, ablation, and computational-efficiency settings. Results: The proposed approach achieves in-dataset classification accuracies of 94.87%, 97.32%, and 98.17% on ISIC 2018, HAM10000, and PH2, respectively. It outperforms strong transformer and hybrid baselines. Cross-dataset transfer experiments show consistent performance gains of 3.5–5.8% over supervised counterparts. This indicates improved robustness to domain shift. Furthermore, when fine-tuned with only 10% of the labeled training data, the model achieves performance comparable to fully supervised baselines. Conclusions: This highlights strong data efficiency. These results demonstrate that dermatology-specific foundation learning offers a principled and practical solution for robust dermoscopic lesion classification under realistic clinical constraints. Full article

Automated anomaly detection in transportation infrastructure is essential for enhancing safety and reducing the operational costs associated with manual inspection protocols. This study presents an improved neuromorphic vision system, which extends the prior SIFT-SNN (scale-invariant feature transform–spiking neural network) proof-of-concept by incorporating temporal feature aggregation for context-aware and sequence-stable detection. Analysis of classical stitching-based pipelines exposed sensitivity to motion and lighting variations, motivating the proposed temporally smoothed neuromorphic design. SIFT keypoints are encoded into latency-based spike trains and classified using a leaky integrate-and-fire (LIF) spiking neural network implemented in PyTorch. Evaluated across three hardware configurations—an NVIDIA RTX 4060 GPU, an Intel i7 CPU, and a simulated Jetson Nano—the system achieved 92.3% accuracy and a macro F1 score of 91.0% under five-fold cross-validation. Inference latencies were measured at 9.5 ms, 26.1 ms, and ~48.3 ms per frame, respectively. Memory footprints were under 290 MB, and power consumption was estimated to be between 5 and 65 W. The classifier distinguishes between safe, partially dislodged, and fully dislodged barrier pins, which are critical failure modes for the Auckland Harbour Bridge’s Movable Concrete Barrier (MCB) system. Temporal smoothing further improves recall for ambiguous cases. By achieving a compact model size (2.9 MB), low-latency inference, and minimal power demands, the proposed framework offers a deployable, interpretable, and energy-efficient alternative to conventional CNN-based inspection tools. Future work will focus on exploring the generalisability and transferability of the work presented, additional input sources, and human–computer interaction paradigms for various deployment infrastructures and advancements. Full article

Background/Objectives: There is limited empirical evidence on the feasibility of inclusive, community-based exercise programs for older adults with long-term mobility disabilities. This pilot study investigated the feasibility and preliminary effectiveness of a community-based high-intensity functional training (HIFT) intervention. Methods: This single-group pre–post feasibility trial was delivered across four community-based HIFT facilities. Thirteen participants enrolled, and 10 (mean age 69.8 ± 6.7 years; 60% female) completed baseline assessments, two onboarding sessions, and thrice-weekly group-based workouts across 16 weeks. Physical function was assessed using the Canadian Occupational Performance Measure (COPM), Patient-Reported Outcomes Measurement Information System (PROMIS) Physical Function, Modified Falls Efficacy Scale (MFES), and standardized tests of mobility, balance, and strength. Exploratory outcomes included body mass index (BMI), waist circumference, work capacity, and quality of life (QOL). Results: Recruitment, retention, and attendance rates were 38%, 77%, and 58% (80% including make-up sessions), respectively. The intervention was safe and well-tolerated, with one fall-related adverse event. Self-reported functional outcomes demonstrated small to large effects, with large improvements in participant-identified functional activities (d = 1.03–1.54) and fall efficacy (d = 0.97), and a small effect for standardized physical function (d = 0.36) Endurance improved substantially (d = 1.01), while mobility, balance, and strength outcomes reflected maintenance or small to moderate gains (d = 0.08–0.55). BMI remained stable (d = 0.05), work capacity increased with moderate to large effects (d = 0.61–1.43), and QOL improved modestly (d = 0.20). Exit interviews reinforced high acceptability, highlighting individualized adaptations, supportive trainers, and the group-based context as motivating contextual factors. Conclusions: A community-based HIFT program is feasible and acceptable for older adults with mobility disabilities. Full article

Autism spectrum disorder (ASD) is characterized by persistent difficulties in communication, language, and social interaction, which requires innovative strategies and resources that promote educational inclusion and personal autonomy. In this context, digital technologies have established themselves as support tools with significant potential for the communicative and linguistic development of people with ASD. The aim of this study was to conduct a systematic review of recent scientific literature on the use of digital applications aimed at developing communication and language in people with ASD. The search was carried out in the Scopus, Google Scholar, and Mendeley databases, covering the period from 2019 to 2024. The methodological criteria of the PRISMA statement were applied, resulting in a total of 61 studies that met the inclusion criteria. The results show that digital applications implemented in educational, family, and community contexts promote linguistic comprehension and expression, increase motivation and active participation, and enhance the functional autonomy of people with ASD. However, limitations were identified related to technological accessibility, specific training for teaching and therapeutic staff, and the scarcity of longitudinal studies assessing the sustained impact of these interventions. In conclusion, this review offers an up-to-date and rigorous synthesis that can guide teachers, therapists, families, and researchers in the selection and use of digital applications as inclusive resources, contributing to the strengthening of communication, social participation, and quality of life for people with ASD. Full article

This study examines teachers’ awareness, experiences, and perceptions concerning Ocean Literacy (OL), with particular emphasis on their knowledge levels and the pedagogical, structural, and institutional challenges associated with integrating OL into formal education. Since the early 2000s, the concept of Ocean Literacy has been defined as understanding the ocean’s influence on humans and humans’ influence on the ocean. The paper draws on work conducted within the Erasmus+ project Sea Tales, one of whose aims was to explore and deepen understanding of Ocean Literacy. Employing a cross-cultural comparative design, this research analyses how OL is conceptualised and enacted within educational contexts in Iceland, Greece, Portugal, and Romania. A mixed-methods methodology was adopted, comprising a systematic literature review, country-specific investigations, co-design sessions with educators and relevant OL stakeholders, and a large-scale teacher survey (N = 266). Findings indicate a pronounced discrepancy between teachers’ high motivation to incorporate OL and the limited institutional and pedagogical support available to them. The study advocates for the development of a holistic, flexible, and multi-tiered teacher training framework that is responsive to contextual constraints, offers accessible and ready-to-use materials, and provides differentiated pathways that cater to both novice and experienced educators. Full article

Accurate gait-phase detection is essential for rehabilitation monitoring, prosthetic control, and human–robot interaction. Artificial intelligence supports continuous, personalized mobility assessment by extracting clinically meaningful patterns from wearable sensors. A richer view of gait dynamics can be achieved by integrating additional signals, including inertial, plantar flex, footswitch, and EMG data, leading to more accurate and informative gait analysis. Motivated by these needs, this study investigates discrete gait-phase recognition for the right leg using a multi-subject IMU dataset collected from lower-limb sensors. IMU recordings were segmented into 128-sample windows across 23 channels, and each window was flattened into a 2944-dimensional feature vector. To ensure reliable ground-truth labels, we developed an automatic relabeling pipeline incorporating heel-strike and toe-off detection, adaptive threshold tuning, and sensor fusion across sensor modalities. These windowed vectors were then used to train a comprehensive suite of machine learning models, including Random Forests, Extra Trees, k-Nearest Neighbors, XGBoost, and LightGBM. All models underwent systematic hyperparameter tuning, and their performance was assessed through k-fold cross-validation. The results demonstrate that tree-based ensemble models provide accurate and stable gait-phase classification with accuracy exceeding 97% across both test sets, underscoring their potential for future real-time gait analysis and lower-limb assistive technologies. Full article

Reliable Internet connectivity is essential for latency-sensitive services such as video conferencing, media streaming, and online gaming. Round-trip time (RTT) is a key indicator of network performance and is central to setting retransmission timeout (RTO); inaccurate RTT estimates may trigger unnecessary retransmissions or slow loss recovery. This paper proposes an Enhanced Regularized Extreme Learning Machine (RELM) for RTT estimation that improves generalization and efficiency by interleaving a bidirectional log-step heuristic to select the regularization constant C. Unlike manual tuning or fixed-range grid search, the proposed heuristic explores C on a logarithmic scale in both directions ($\times 10$ and /10) within a single loop and terminates using a tolerance–patience criterion, reducing redundant evaluations without requiring predefined bounds. A custom RTT dataset is generated using Mininet with a dumbbell topology under controlled delay injections (1–1000 ms), yielding 1000 supervised samples derived from 100,000 raw RTT measurements. Experiments follow a strict train/validation/test split (6:1:3) with training-only standardization/normalization and validation-only hyperparameter selection. On the controlled Mininet dataset, the best configuration (ReLU, 150 hidden neurons, $C={10}^2$) achieves $R^2=0.9999$, $MAPE=0.0018$, $MAE=966.04$, and $RMSE=1589.64$ on the test set, while maintaining millisecond-level runtime. Under the same evaluation pipeline, the proposed method demonstrates competitive performance compared to common regression baselines (SVR, GAM, Decision Tree, KNN, Random Forest, GBDT, and ELM), while maintaining lower computational overhead within the controlled simulation setting. To assess practical robustness, an additional evaluation on a public real-world WiFi RSS–RTT dataset shows near-meter accuracy in LOS and mixed LOS/NLOS scenarios, while performance degrades markedly under dominant NLOS conditions, reflecting physical-channel limitations rather than model instability. These results demonstrate the feasibility of the Enhanced RELM and motivate further validation on operational networks with packet loss, jitter, and path variability. Full article

Dogs show exceptional olfactory sensitivity and are widely used in medical, rescue, military, and forensic applications, yet the determinants of individual and breed-level scent-work performance remain incompletely characterized. This review integrates evidence from the anatomy and physiology of the canine olfactory organ, neurobiological mechanisms of odor transduction and coding, and links between olfaction, memory, and emotion, alongside molecular genetics, evolution, domestication, and selective breeding. We synthesize findings indicating that complex nasal turbinates and specialized airflow patterns enhance odorant capture, while olfactory bulb circuitry and downstream connections to limbic and frontal networks support discrimination, learning, and affective modulation. Comparative and breed-focused studies suggest that skull morphology and breeding priorities can alter olfactory capacity, with shortened nasal anatomy associated with reduced functional potential in some lines. In applied contexts, detection success is strongly shaped by behavioral traits such as motivation, persistence, independence, and reward value, as well as by physical condition and environmental stressors that can impair search efficiency. Emerging literature further suggests that the gastrointestinal and upper airway microbiome, together with diet, housing, temperature, and workload, may influence sensory and cognitive readiness, although direct causal links to detection outcomes remain limited. Overall, canine olfactory performance reflects interactions among genetic–anatomical capacity, neurobehavioral factors, and environment, underscoring the value of standardized selection, training, welfare management, and future integrative research. Full article