Search Results (141)

Mine water hazards represent one of the principal threats to safe coal mine operations; therefore, accurately predicting mine water inflow is critical for drainage system design and water hazard mitigation. Because mine water inflow is governed by the combined influence of multiple hydrogeological factors and thus exhibits pronounced non-linear characteristics, conventional approaches are inadequate in terms of forecasting accuracy and medium- to long-term predictive capability. To address this issue, this study proposes a Chaos-Autoformer-based method for predicting mine water inflow. First, the univariate inflow series is mapped into an m-dimensional phase space by means of phase-space reconstruction from chaos theory, thereby fully preserving its non-linear features; the reconstructed vectors are then used to train and forecast inflow with an improved Chaos-Autoformer model. On top of the original Autoformer architecture, the proposed model incorporates a Chaos-Attention mechanism and a Lyap-Dropout scheme, which enhance sensitivity to small perturbations in initial conditions and complex non-linear propagation paths while improving stability in long-horizon forecasting. In addition, the loss function integrates the maximum Lyapunov exponent error and earth mode decomposition (EMD) indices so as to jointly evaluate dynamical consistency and predictive performance. An empirical analysis based on monitoring data from the Liangshuijing Coal Mine for 2022–2025 demonstrates that the trained model delivers high accuracy and stable performance. Ablation experiments further confirm the significant contribution of the chaos-aware components: when these modules are removed, forecasting accuracy declines to only 76.5%. Using the trained model to predict mine water inflow for the period from June 2024 to June 2025 yields a root mean square error (RMSE) of 30.73 m³/h and a coefficient of determination (R²) of 0.895 against observed data, indicating excellent fitting and predictive capability for medium- to long-term tasks. Extending the forecast to July 2025–November 2027 reveals a pronounced annual cyclical pattern in future mine water inflow, with markedly higher inflow in summer than in winter and an overall slowly declining trend. These findings show that the Chaos-Autoformer can achieve high-precision medium- and long-term predictions of mine water inflow, thereby providing technical support for proactive deployment and refined management of mine water hazard prevention. Full article

Metastatic cancer, characterized by poor survival outcomes and grim prognosis, represents the final stage of malignancy. The current evidence regarding the prophylactic effects of glucagon-like peptide-1 (GLP-1) receptor agonists and sodium–glucose cotransporter 2 (SGLT2) inhibitors on metastatic cancer remains largely unexamined. With a confirmatory approach based on the Cochrane recommendation, we conducted a frequentist-based network meta-analysis (NMA) of randomized controlled trials (RCTs) evaluating such medications. The primary outcome was the incidence of metastatic cancer, while secondary outcomes included safety profiles assessed through dropout rates. The findings were reaffirmed by sensitivity analysis with a Bayesian-based NMA. This NMA of 207,606 participants from 67 RCTs revealed that only efpeglenatide demonstrated a statistically significant reduction in metastatic cancer events compared to controls (odds ratio = 0.26, 95% confidence intervals = 0.09 to 0.70, p = 0.010, number needed to treat = 188.4). Efpeglenatide’s efficacy was not confined to specific cancer types. Safety profiles were comparable across all treatments. These findings indicate that efpeglenatide may possess a broad, systemic preventive effect against metastatic cancers, potentially operating through mechanisms that are not restricted to individual organ systems. Further research is warranted to elucidate the molecular pathways underlying its anti-metastatic properties and to explore its role in preventive oncology. Full article

Graph convolutional networks (GCNs) have demonstrated remarkable effectiveness in processing graph-structured data and have been widely adopted across various domains. Existing methods mitigate over-smoothing through selective aggregation strategies such as attention mechanisms, edge dropout, and neighbor sampling. While some approaches incorporate global structural context, they often underexplore category-aware representations and inter-category differences, which are crucial for enhancing node discriminability. To address these limitations, a novel framework, CA-NodeNet, is proposed for semi-supervised node classification. CA-NodeNet comprises three key components: (1) coarse-grained node feature learning, (2) category-decoupled multi-branch attention, and (3) inter-category difference feature learning. Initially, a GCN-based encoder is employed to aggregate neighborhood information and learn coarse-grained representations. Subsequently, the category-decoupled multi-branch attention module employs a hierarchical multi-branch architecture, in which each branch incorporates category-specific attention mechanisms to project coarse-grained features into disentangled semantic subspaces. Furthermore, a layer-wise intermediate supervision strategy is adopted to facilitate the learning of discriminative category-specific features within each branch. To further enhance node feature discriminability, we introduce an inter-category difference feature learning module. This module first encodes pairwise differences between the category-specific features obtained from the previous stage and then integrates complementary information across multiple feature pairs to refine node representations. Finally, we design a dual-component optimization function that synergistically combines intermediate supervision loss with the final classification objective, encouraging the network to learn robust and fine-grained node representations. Extensive experiments on multiple real-world benchmark datasets demonstrate the superior performance of CA-NodeNet over existing state-of-the-art methods. Ablation studies further validate the effectiveness of each module in contributing to overall performance gains. Full article

This study explores dropout from professional training programs for early childhood educarers from the perspective of those who left a national training course before completion—a viewpoint previously unexamined in Israel or internationally. While dropout has been studied in various educational contexts, this study addresses the research gap by focusing on the participants themselves. Based on semi-structured interviews with 15 educarers from four training colleges, the study identifies four key themes: (1) the challenge of learning in heterogeneous groups combining beginners and veterans that overlook prior experience; (2) the physical and emotional strain of attending evening classes after full workdays, particularly for mothers of young children; (3) disappointment with the lack of practical tools and an overly theoretical curriculum; and (4) the absence of emotional and professional support mechanisms during the course. These findings highlight a systemic misalignment between the structure of the training and participants’ lived realities. The study expands the job demands–resources model by framing dropout as an outcome of imbalanced responsibility across participants, training institutions, and workplaces. It calls for differentiated, context-sensitive training that integrates practical tools with theoretical content and provides sustained support to strengthen professional capacity and retention in the early childhood workforce. Full article

Background/Objectives: Results from a well-designed trial provide evidence to support approval of truly effective treatments or discontinuation of ineffective treatments. However, the information available at the time of trial design may be limited which may lead to underpowered trials. This work aims to evaluate the impact of design assumption misspecifications on the statistical power of randomized trials with survival outcomes. Methods: The impact of the design assumption misspecifications on statistical power of four different statistical methods was investigated in a simulation study. The methods include the log-rank test, MaxCombo test, the test of difference in survival probability, and test of difference in restricted mean survival time (RMST). The deviations considered include the survival rate in the control arm, the expected treatment effect in terms of magnitude and pattern, accrual rate, and drop-out rate. Results: Deviations in the control arm’s survival distribution have no impact on the power of the log-rank and MaxCombo tests but it affects the trial duration since trials designed with these tests require the total number of events to be met before the final analysis can be conducted. Misspecified treatment effect has similar effect on the statistical power of all four methods. When the proportional hazards assumption is misspecified, the RMST is more robust with a larger early treatment effect, while the survival probability and the MaxCombo tests are more robust with a larger late treatment effect and crossing hazards. Conclusions: Selecting the appropriate statistical tests to design a trial depends on the goal of the trial, the mechanism of action of the experimental treatment, the survival quantity of clinical interest, and the pattern of the expected treatment effect. The final design should be based on assumptions that are as accurate as possible, and the potential impacts of deviations from these assumptions on the trial’s statistical power should be carefully considered. Full article

With the increasing uncertainty introduced by the large-scale integration of renewable energy sources, traditional power dispatching methods face significant challenges, including severe frequency fluctuations, substantial forecasting deviations, and the difficulty of balancing economic efficiency with system stability. To address these issues, a deep learning-based dispatching framework is proposed, which integrates spatiotemporal feature extraction with a stability-aware mechanism. A joint forecasting model is constructed using Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM) to handle multi-source inputs, while a reinforcement learning-based stability-aware scheduler is developed to manage dynamic system responses. In addition, an uncertainty modeling mechanism combining Dropout and Bayesian networks is incorporated to enhance dispatch robustness. Experiments conducted on real-world power grid and renewable generation datasets demonstrate that the proposed forecasting module achieves approximately a 2.1% improvement in accuracy compared with Autoformer and reduces Mean Absolute Error (MAE) and Root Mean Square Error (RMSE) by 18.1% and 14.1%, respectively, compared with traditional LSTM models. The achieved Mean Absolute Percentage Error (MAPE) of 5.82% outperforms all baseline models. In terms of scheduling performance, the proposed method reduces the total operating cost by 5.8% relative to Autoformer, decreases the frequency deviation from 0.158 Hz to 0.129 Hz, and increases the Critical Clearing Time (CCT) to 2.74 s, significantly enhancing dynamic system stability. Ablation studies reveal that removing the uncertainty modeling module increases the frequency deviation to 0.153 Hz and raises operational costs by approximately 6.9%, confirming the critical role of this module in maintaining robustness. Furthermore, under diverse load profiles and meteorological disturbances, the proposed method maintains stable forecasting accuracy and scheduling policy outputs, demonstrating strong generalization capabilities. Overall, the proposed approach achieves a well-balanced performance in terms of forecasting precision, system stability, and economic efficiency in power grids with high renewable energy penetration, indicating substantial potential for practical deployment and further research. Full article

Carbapenem resistance is considered one of the greatest current threats to public health, particularly in the management of infections in clinical settings. Carbapenem resistance in bacteria is mainly due to mechanisms such as the production of carbapenemases (such as the imipenemase IMP, or other enzymes like VIM, NDM, and KPC), that can be detected by several laboratory tests, including immunochromatography and automated real-time PCR (qPCR). Methods: As part of local studies to monitor carbapenem-resistant bacteria in Costa Rica, two cases were initially identified with inconsistent IMP detection results. A possible gene drop-out in the automated qPCR test was suggested based on the negative result, contrasting with the positive result by immunochromatography and whole-genome sequencing. We hypothesized that molecular testing could be optimized through the development of tailored assays to improve the detection of IMP genes. Thus, using IMP gene sequences from the local isolates and regional sequences in databases, primers were redesigned to extend the detection of IMP alleles of regional relevance. Results: The tailored qPCR was applied to a local collection of 119 carbapenem-resistant isolates. The genomes of all 14 positive cases were sequenced, verifying the results of the custom qPCR, despite the negative results of the automated testing. Conclusions: Guided by whole-genome sequencing, it was possible to extend the molecular detection of IMP alleles circulating in Latin America using a tailored qPCR to overcome IMP gene drop-out and false-negative results in an automated qPCR. Full article

In the task of PolSAR image classification, effectively utilizing convolutional neural networks and vision transformer models with limited labeled data poses a critical challenge. This article proposes a novel method for PolSAR image classification that combines channel-wise convolutional local perception, detachable self-attention, and a residual feedforward network. Specifically, the proposed method comprises several key modules. In the channel-wise convolutional local perception module, channel-wise convolution operations enable accurate extraction of local features from different channels of PolSAR images. The local residual connections further enhance these extracted features, providing more discriminative information for subsequent processing. Additionally, the detachable self-attention mechanism plays a pivotal role: it facilitates effective interaction between local and global information, enabling the model to comprehensively perceive features across different scales, thereby improving classification accuracy and robustness. Subsequently, replacing the conventional feedforward network with a residual feedforward network that incorporates residual structures aids the model in better representing local features, further enhances the capability of cross-layer gradient propagation, and effectively alleviates the problem of vanishing gradients during the training of deep networks. In the final classification stage, two fully connected layers with dropout prevent overfitting, while softmax generates predictions. The proposed method was validated on the AIRSAR Flevoland, RADARSAT-2 San Francisco, and RADARSAT-2 Xi’an datasets. The experimental results demonstrate that the proposed method can attain a high level of classification performance even with a limited amount of labeled data, and the model is relatively stable. Furthermore, the proposed method has lower computational costs than comparative methods. Full article

Onion cultivation in the Republic of Korea is increasingly threatened by labor shortages and an aging rural population, underscoring the growing importance of mechanization. This study evaluated the combined and individual performances and economic feasibility of mechanized transplanting, stem cutting, harvesting, and collecting operations using work efficiency; the missing plant, stem cutting, damage, and dropout rates; and foreign matter content as indicators. Mechanized operations achieved up to 358-fold higher work efficiencies than manual labor operations. However, in terms of marketability, performance was inferior due to missing plants, improperly cut stems, damaged bulbs, dropped onions, and foreign matter contamination. The economic analysis indicated that the use of individual machines is advantageous for farms larger than 10.2 ha for transplanting, 1.14 ha for stem cutting, 0 ha for harvesting (i.e., profitable regardless of farm size), and 6.95 ha for collecting. For fully mechanized operations, using machines for all four processes, the break-even area was found to be 3.63 ha, with a payback period of 2.1 years. These findings are expected to serve as a foundational reference for onion growers considering the adoption of mechanization. Full article

Rolling bearings are indispensable but also the most fault-prone components of rotating machinery, typically used in fields such as industrial aircraft, production workshops, and manufacturing. They encounter diverse mechanical stresses, such as vibration and friction during operation, which may lead to wear and fatigue cracks. From this standpoint, the present study combined a 1-D convolutional neural network (1-D CNN) with a long short-term memory (LSTM) algorithm for classifying different ball-bearing health conditions. A physics-guided method that adopts fault characteristics frequencies was used to calculate an optimal input size (sample length). Moreover, grid search was utilized to optimize (1) the number of epochs, (2) batch size, and (3) dropout ratio and further enhance the efficacy of the proposed 1-D CNN-LSTM network. Therefore, an attempt was made to reduce epistemic uncertainty that arises due to not knowing the best possible hyper-parameter configuration. Ultimately, the effectiveness of the physics-guided optimized 1-D CNN-LSTM was tested by comparing its performance with other state-of-the-art models. The findings revealed that the average accuracies could be enhanced by up to 20.717% with the help of the proposed approach after testing it on two benchmark datasets. Full article

Accurately forecasting karst spring discharge remains a significant challenge due to the inherent nonstationarity and multi-scale hydrological dynamics of karst hydrological systems. This study presents a physics-informed variational mode decomposition long short-term memory (VMD-LSTM) model, enhanced with an attention mechanism and Monte Carlo dropout for uncertainty quantification. Hourly discharge data (2013–2018) from the Zhaidi karst spring in southern China were decomposed using VMD to extract physically interpretable temporal modes. These decomposed modes, alongside precipitation data, were input into an attention-augmented LSTM incorporating physics-informed constraints. The model was rigorously evaluated against a baseline standalone LSTM using an 80% training, 15% validation, and 5% testing data partitioning strategy. The results demonstrate substantial improvements in prediction accuracy for the proposed framework compared to the standard LSTM model. Compared to the baseline LSTM, the RMSE during testing decreased dramatically from 0.726 to 0.220, and the NSE improved from 0.867 to 0.988. The performance gains were most significant during periods of rapid conduit flow (the peak RMSE decreased by 67%) and prolonged recession phases. Additionally, Monte Carlo dropout, using 100 stochastic realizations, effectively quantified predictive uncertainty, achieving over 96% coverage in the 95% confidence interval (CI). The developed framework provides robust, accurate, and reliable predictions under complex hydrological conditions, highlighting substantial potential for supporting karst groundwater resource management and enhancing flood early-warning capabilities. Full article

Many vocational education and training (VET) students in the Netherlands struggle with the transition from education to the labour market, resulting in high dropout rates. VET institutions are actively seeking effective mechanisms to support their students during this transition. This research explored valuable strategies identified by education professionals to facilitate a sustainable transition from education to the labour market. This study employed a realistic evaluation framework using CIMO-logic (which focuses on the Context, Intervention, Mechanism, Outcome) for analysis in order to gain insight into the processes of change. In total, four cases were studied at two Dutch educational institutions. The research followed an inductive approach using within-case and cross-case analyses. Five key elements were identified: skills and competencies, Self-insight, Self-efficacy, Building a professional network, and bridging education and practice. In school-to-work guidance for VET students, these elements are relevant to consider in guidance programs. Full article

In the swimming context, practitioners show difficulties in learning its basic skills, and the emotional factor seems to be one of the triggers for these complications, with “fear” standing out as one of the most studied emotions due to its cognitive reactive nature associated with survival mechanisms. This emotional response can hinder the learning process in swimming, potentially leading to disengagement or dropout. The present study aimed to analyze the causes that lead to fear of the aquatic environment, its effects on learning swimming, and how swimming coaches can intervene to help overcome this fear. Direct observation was used to capture the individuals’ perception of the degree of fear. Subsequently, semi-structured interviews were conducted to analyze an intervention aimed at reducing the fear of water, followed by a corresponding content analysis. The fear of water is commonly associated with anxiety, panic, and muscle tension. The role of the swimming instructor is crucial, as their teaching approach significantly influences the swimmer’s emotional response, particularly in fostering a sense of security. The use of playful activities proves effective in helping children adapt, overcoming the limitations posed by the fear of water. Recognizing students’ fears allows instructors to structure swimming lessons effectively, helping students overcome their emotional barriers. Therefore, introducing children to the aquatic environment at an early age contributes to this goal. Full article

The development of Facial Expression Recognition (FER) technology has significantly enhanced the naturalness and intuitiveness of human-robot interaction. In the field of service robots, particularly in applications such as production assistance, caregiving, and daily service communication, efficient FER capabilities are crucial. However, existing Convolutional Neural Network (CNN) models still have limitations in terms of feature representation and recognition accuracy for facial expressions. To address these challenges, we propose CAGNet, a novel network that combines multiscale feature aggregation and attention mechanisms. CAGNet employs a deep learning-based hierarchical convolutional architecture, enhancing the extraction of features at multiple scales through stacked convolutional layers. The network integrates the Convolutional Block Attention Module (CBAM) and Global Average Pooling (GAP) modules to optimize the capture of both local and global features. Additionally, Batch Normalization (BN) layers and Dropout techniques are incorporated to improve model stability and generalization. CAGNet was evaluated on two standard datasets, FER2013 and CK+, and the experiment results demonstrate that the network achieves accuracies of 71.52% and 97.97%, respectively, in FER. These results not only validate the effectiveness and superiority of our approach but also provide a new technical solution for FER. Furthermore, CAGNet offers robust support for the intelligent upgrade of service robots. Full article

Estimating Remaining Useful Life (RUL) is crucial in modern Prognostic and Health Management (PHM) systems providing valuable information for planning the maintenance strategy of critical components in complex systems such as aircraft engines. Deep Learning (DL) models have shown great performance in the accurate prediction of RUL, building hierarchical representations by the stacking of multiple explicit neural layers. In the current research paper, we follow a different approach presenting a Deep Equilibrium Model (DEM) that effectively captures the spatial and temporal information of the sequential sensor. The DEM, which incorporates convolutional layers and a novel dual-input interconnection mechanism to capture sensor information effectively, estimates the degradation representation implicitly as the equilibrium solution of an equation, rather than explicitly computing it through multiple layer passes. The convergence representation of the DEM is estimated by a fixed-point equation solver while the computation of the gradients in the backward pass is made using the Implicit Function Theorem (IFT). The Monte Carlo Dropout (MCD) technique under calibration is the final key component of the framework that enhances regularization and performance providing a confidence interval for each prediction, contributing to a more robust and reliable outcome. Simulation experiments on the widely used NASA Turbofan Jet Engine Data Set show consistent improvements, with the proposed framework offering a competitive alternative for RUL prediction under diverse conditions. Full article