Search Results (6,063)

Walking is not merely locomotion but a window into the nervous system, integrating cortical, subcortical, cerebellar, spinal, and peripheral networks into a unified motor behavior. Across neurological diseases—including Parkinson’s disease, atypical parkinsonism, cerebellar ataxias, stroke, multiple sclerosis, neuropathies, neuromuscular disorders, and functional gait syndromes—gait disturbances are among the most disabling clinical features, contributing to falls, loss of independence, institutionalization, and premature mortality. Traditional bedside observation remains indispensable, but it lacks the sensitivity and reproducibility needed to capture subtle, episodic, or prodromal abnormalities. Over the past decade, advances in wearable sensors, marker-based and markerless motion capture, pressure-sensitive walkways, force plates, artificial intelligence, and machine learning have positioned digital mobility outcomes as promising, ecologically valid biomarkers of neurological function. These measures can support differential diagnosis, provide prognostic information on falls and survival, and serve as sensitive endpoints in therapeutic trials. They may also detect early abnormalities, such as increased stride-to-stride variability or prolonged double-support time, before overt clinical deterioration becomes evident. Clinical applications are increasingly evident across disorders, including distinguishing Parkinson’s disease from atypical parkinsonism, quantifying treatment response in normal-pressure hydrocephalus, tracking progression in ataxia and multiple sclerosis, predicting functional decline in motor neuron disease, and guiding rehabilitation after stroke. Integration with neuroimaging, electrophysiology, and molecular biomarkers is beginning to reveal the circuits underlying variability, instability, and freezing, positioning gait as a systems-level marker of neural integrity. Nevertheless, methodological heterogeneity, limited disease-specific validation, insufficient longitudinal data, and lack of consensus on clinically meaningful parameters continue to constrain translation. Cognitive, affective, and environmental influences also remain insufficiently represented in digital frameworks, while equity, accessibility, algorithmic bias, and privacy require careful ethical governance. Reconceptualizing gait as a “sixth vital sign” reframes mobility as a multidimensional biomarker of neural and systemic health. With harmonized protocols, robust validation, multimodal integration, and appropriate ethical frameworks, gait analysis could become a cornerstone of precision neurology. Full article

School buses serve as the primary mode of transportation for children traveling to and from school, and their safety measures represent a critical safeguard for children’s lives. Nevertheless, incidents in which children are left unattended on school buses—due to inadequate supervision or the children’s own actions—occur with notable frequency and can lead to fatal outcomes. To mitigate or prevent such tragedies, this paper proposes an in-vehicle thermal imaging solution based on infrared array sensors, integrated with a dedicated algorithm to detect whether a child has been left behind in the school bus. The system collects background temperature, presence temperature, and real-time temperature data inside the bus using infrared array sensors. By comparing the real-time temperature difference against a predefined presence temperature difference threshold, the algorithm determines whether a child is present under the current thermal conditions. It then verifies whether the number of positive detections within a specified temperature range meets a preset presence count threshold, thereby reaching a final decision regarding child presence. Experiments identified optimal parameters: a temperature range of 26–33 °C, a double-difference threshold (ε = 1), and a presence count threshold (P = 4). Random testing demonstrated that the proposed technical solution and algorithm achieve an overall detection success rate of 92.5%. This study develops a low-cost, easily deployable, non-contact thermal imaging method capable of identifying forgotten children on school buses with satisfactory accuracy. By detecting retention before harm occurs, the approach enhances the safety of children traveling by school bus. Full article

Offshore wind energy offers abundant resources and significant potential for large-scale development. Efficient design of collection systems is critical to the economic viability of offshore wind farms (OWFs). This study proposes an upper–lower level topology optimization framework based on the Artificial Lemming Algorithm (ALA) to address the complexity arising from large numbers of wind turbines (WTs). At the upper level, wind turbines can be partitioned into different numbers of regions according to practical engineering requirements using the Radial Fuzzy C-Means (RFCM) clustering algorithm. At the lower level, the ALA is applied to optimize the collection system topology within each region, aiming to minimize total construction cost while satisfying operational constraints. A case study involving a 75-WT offshore wind farm is conducted. Comparative simulations against various heuristic algorithms including Particle Swarm Optimization (PSO), Genetic Algorithm (GA), and Differential Evolution (DE) show that the proposed method achieves faster convergence, lower total costs and greater robustness. Specifically, the ALA reduces the best cost by 9.9% and improves average runtime by 28.5%, indicating its advantages in best-cost search and computational efficiency in the tested case. In addition, based on 10 independent runs, the ALA achieves the lowest median cost of $6684\times {10}^4$ CNY, with an interquartile range of 6593–$6813\times {10}^4$ CNY and a cost range of 6362–$7087\times {10}^4$ CNY. Overall, the proposed framework provides a practical optimization approach for obtaining low-cost feasible collection-system layouts in the studied offshore wind farm case. Full article

Background: A subgroup of adolescent and young adult patients (AYAs; 18 to 39 years at diagnosis) face an uncertain or poor cancer prognosis (UPCP). Previous qualitative research identified dual coping pathways in this population: engagement in life versus the reality of premature death. This study examines whether similar psychosocial profiles can be identified through quantitative data, aiming to differentiate patient experiences and identify characteristic features of each cluster. Additionally, this study examines the association between cluster membership and social support needs to understand psychosocial disparities. Methods: Eligible participants completed questionnaires assessing physical, psychosocial, and existential outcomes related to their disease and prognosis. An ensemble clustering approach was applied, including evaluation of clustering tendency and multiple algorithms, with stable clusters identified through majority voting. Associations with social support needs were analyzed using Fisher’s exact test. Results: Data from 155 AYAs with a UPCP were included. The mean age at diagnosis was 31.2 years, with glioma (34.8%) and breast cancer (17.4%) as the most common diagnoses. Two distinct clusters were identified: one (22%) characterized by poorer functional outcomes and fewer protective factors (e.g., hope, meaning in life), and another cluster (78%) with better functioning and less frequent needs for social support (p < 0.00043). Conclusions: Findings revealed divergent psychosocial profiles within the AYA-UPCP population, highlighting the importance of early identification of vulnerable subgroups. Strengthening protective factors may enhance resilience and reduce unmet support needs. Validation in larger, external datasets is needed to confirm these pathways and guide tailored supportive care strategies. Full article

Background: Some widely used wildland fire behavior models, like the Fire Area Simulator (FARSITE), propagate fire fronts by computing the front-normal velocity (spread rate) as a function of local inputs and the front-normal direction. Such models are sometimes observed to cause the collapse of crown fires into sharp wedge shapes that eliminate heading fire behavior. Aims: We set out to document this phenomenon and, more generally, understand the relationships between fire shapes and spread rate functions. Methods: The phenomenon is studied both mathematically and through simulation experiments. Non-smooth fire fronts are theorized mathematically by an Eikonal partial differential equation ($H(x,\tau ,D\tau )=1$), where the unknown $\tau \left(x\right)$ is the time-of-arrival function and the Hamiltonian $H(x,t,p)$ is positively homogeneous and possibly non-convex in p; convex analysis is used to study viscosity solutions in constant conditions. Results: We show that a fire spread model preserves the smoothness of fire fronts if and only if it is equivalent to using the Huygens principle. Nontrivially, this is equivalent to a convexity criterion on the inverse spread rate profile, which is then the polar dual of the Huygens wavelet; this corresponds to Hamiltonian–Lagrangian duality. The relevance of smoothness-destroying models to crown fire is debated. Exact analytical formulas are derived for fire growth in constant conditions. Conclusions: Our understanding of fire spread models is improved by solving the spread equations in more general ways than previously known. In particular, the collapse of heading crown fires into sharp shapes is now explained. Smoothness-destroying spread models cannot be simulated by algorithms based on travel time like cellular automata; their general well-definedness remains an open question. Fire modelers can use these findings to guide their search for improved crown fire models, and more generally to verify the accuracy of numerical implementations. Full article

In the task planning of heterogeneous unmanned aerial vehicle formations, problems such as dynamic topological instability and sparse Pareto front exist, which affect the robustness of the planning. To address this, this paper proposes a cooperative task planning method based on multi-objective dolphin echolocation optimization driving. Firstly, a differentiated dynamic model of heterogeneous unmanned aerial vehicles covering different configurations such as rotors and fixed wings is constructed, and a dynamic communication topology model is established based on time-varying graph theory to quantify transmission delay and link stability. Then, a multi-objective optimization model is designed with task completion, energy balance, and time cost as the core, Bayesian networks are introduced to construct a dynamic threat field, and risk assessment and real-time response are achieved in complex environments. Based on this, a multi-objective dolphin echo optimization algorithm is adopted to solve the model, and its echo beam focusing search and adaptive weight allocation mechanism are utilized to effectively improve the convergence and distribution of the Pareto solution set. Finally, a “decision execution” hierarchical collaborative control architecture is constructed, utilizing the decision layer to output a global planning scheme and the execution layer to achieve rolling optimization and precise tracking of instructions through distributed model predictive control. The simulation test results show that this method can maintain high task completion, energy balance, and communication stability in different formation sizes and complex environments significantly better than traditional algorithms. When the formation size is between 20 and 60 sorties, the hypervolume (HV) index of this method is superior to that of the comparison method. In cases of sudden obstacles and complex electromagnetic interference scenarios, the average energy consumption of a single unmanned aerial vehicle after applying this method is maintained at 150–250 Wh, and the transmission delay is stable at 50–200 ms. The experimental results verify that this method has good planning robustness and collaborative real-time performance. Full article

This paper addresses the imaging crosstalk problem in 2D spectra from the LAMOST Phase II upgrade, caused by increased fiber density. We propose HyperDecouple_Net, a hypernetwork-based decoupling algorithm designed to overcome key limitations of existing deep learning models, including overlapping-layer collapse and structural distortion. The method integrates an adaptive overlapping-layer enhancement module, a dual-scale hypernetwork differential decoupling module, and a linear consistency constraint module. Additionally, we introduce LAMOST-SD-2026, a public dataset comprising 15,500 linearly superimposed spectral samples with ground-truth labels, derived from real LAMOST Phase I observations. Experimental results on this dataset show that HyperDecouple_Net achieves superior performance, with a PSNR_A of 12.71 dB, PSNR_B of 10.87 dB, SSIM_B of 0.3895, and SAM of 0.4841, outperforming both traditional methods (e.g., NMF, ICA) and recent deep learning approaches. The proposed method can be directly integrated into the LAMOST Phase II preprocessing pipeline, offering a robust solution for high-precision spectral decoupling and supporting the scientific output of the survey. Full article

The growing worldwide prevalence of Diabetes Mellitus highlights the urgent need for effective early detection methods to enable prompt intervention. This study develops a machine learning-based decision-support prototype for predicting diabetes risk using health metrics from the DiaHealth dataset, a recently published Bangladeshi open-source dataset for Type 2 diabetes prediction. Five supervised learning algorithms were evaluated: Logistic Regression (LR), Support Vector Machine (SVM), K-Nearest Neighbour (KNN), Decision Tree (DT), and Random Forest (RF). Models were assessed across three stages: before feature scaling, after standardisation, and following hyperparameter optimisation via GridSearchCV, using accuracy, precision, recall, and F1-score as evaluation metrics. LR and SVM showed marked improvements after standardisation, consistent with their sensitivity to feature magnitude, whilst tree-based approaches such as DT and RF remained largely unchanged. KNN displayed minimal sensitivity to scaling, which is discussed in relation to the feature distributions of the dataset. Following hyperparameter tuning, RF achieved the highest accuracy of 95%, outperforming all other models. RF predictions were interpreted using Local Interpretable Model-agnostic Explanations (LIME) to promote transparency in model decision-making. The best-performing model was subsequently deployed as an interactive web-based prototype application using Streamlit, providing real-time prediction outputs. These findings demonstrate how preprocessing choices and hyperparameter tuning can differentially affect algorithm performance and illustrate the potential of combining explainable AI with practical deployment for diabetes risk assessment in a research context. Full article

The large-scale integration of wind and solar power introduces significant volatility into electricity markets, posing challenges for accurate day-ahead price forecasting for generation companies. This paper proposes a hybrid forecasting model, CEEMD-SE-IBA-LSTM, based on hyperparameter adaptation to improve prediction accuracy. First, a similar-day selection method integrating Random Forest and an Improved Grey Ideal Value approximation identifies the most relevant historical days. Second, Complete Ensemble Empirical Mode Decomposition with Sample Entropy (CEEMD-SE) decomposes and reconstructs the price series into stable components. Third, an Improved Bat Algorithm (IBA), incorporating differential evolution and adaptive weighting, is developed to optimize two key LSTM hyperparameters: the number of hidden layer neurons, which is treated as a model architecture hyperparameter, and the learning rate, which is treated as a training hyperparameter. The number of LSTM layers and the number of training epochs are kept fixed as model settings to ensure reproducibility. Using data from the US PJM market, the proposed model is validated against six benchmarks. The results show that CEEMD-SE-IBA-LSTM achieves superior performance, with a Mean Absolute Percentage Error (MAPE) of 3.73%, a Root Mean Square Error (RMSE) of 3.57 $/MWh, and a Mean Absolute Error (MAE) of 1.95 $/MWh. The method provides accurate price trends, offering effective decision support for new energy enterprises in price bidding to enhance revenue. Full article

In the context of an increasingly complex and dynamic digital ecosystem, accurately assessing the value of Specialized, Refined, Differentiated, and Innovative (SRDI) enterprises is crucial for making effective decisions. Traditional valuation methods struggle to effectively address issues such as the high R&D expenditures and significant operational risks associated with these enterprises. This study proposes an interpretable intelligent decision-support framework for valuing SRDI enterprises listed on the Beijing Stock Exchange (BSE), constructing a multidimensional indicator system that encompasses solvency, profitability, and R&D capabilities. Feature importance screening using the XGBoost algorithm was conducted to identify key indicators as input variables for a backpropagation (BP) neural network. Concurrently, the Particle Swarm Optimization (PSO) algorithm was applied to the neural network to optimize initial weights and thresholds, thereby modeling nonlinear valuation relationships. Empirical analysis of 770 SRDI firms listed on the Beijing Stock Exchange from 2020 to 2024 indicates that the XGBoost-PSO-BPNN model achieved a coefficient of determination of 0.8083 on the test set, outperforming traditional linear models and benchmark models such as single-tree models. SHAP explainability analysis further reveals that current asset turnover, return on assets, and equity concentration are the primary value drivers. This study employs various clustering methods to further classify enterprises into three categories and proposes recommendations for differentiated regulatory policies, providing intelligent decision support for enterprises operating within complex digital ecosystems. Full article

Introduction: Adolescent depression and anxiety are major public health concerns. Previous studies showed that low mental health literacy is associated with depressive and anxiety symptoms. However, how its core dimensions—knowledge, attitudes, and skills—differentially relate to emotional symptoms remains unclear. Methods: A school-based survey was conducted among 6400 adolescents in Guangdong, China. Eligible participants completed the MHL questionnaire and assessments for depressive and anxiety symptoms. We assessed whether MHL was associated with depressive and anxiety symptoms in adolescents. Machine learning algorithms with SHAP analysis were applied to explore complex associations and validate key findings. Results: A total of 5759 adolescents were included. MHL and the knowledge dimension were negatively associated with depressive and anxiety symptoms. The attitudes dimension showed a negative association with both mental health outcomes (depression: OR = 0.83; anxiety: OR = 0.84) and machine learning confirmed attitudes as the key factor. Skills were unrelated to depressive symptoms. At the highest quartile, skills showed a positive association with anxiety symptoms (OR = 1.29). Conclusions: The attitudes dimension is negatively associated with adolescent depressive and anxiety symptoms and emerged as a key feature in ML identification models. Full article

Accurate electricity price forecasting (EPF) is essential for market participants to optimize trading strategies and for power systems to accommodate the increasing penetration of volatile renewable energy sources. However, electricity price series are characterized by strong nonlinearity, high volatility, and significant structural breaks, which pose substantial challenges to conventional forecasting models. Although numerous hybrid deep learning models have been proposed for EPF, most existing approaches either overlook structural breaks or treat them as outliers rather than as signals of regime shifts, often resulting in systematic forecasting degradation when market conditions change abruptly. To address this issue, this study proposes COCAL-TTL, a novel multi-stage structural break-aware forecasting framework that integrates regime-adaptive data partitioning with a functionally differentiated hybrid deep learning architecture. First, a joint detection scheme combining the Iterated Cumulative Sum of Squares (ICSS) algorithm and the Chow test is employed to partition Spanish electricity market data from 2014 to 2023 into distinct regimes. Within each regime, CEEMDAN is applied to extract multi-scale features, which are subsequently reconstructed into trend, periodic, and random components based on an independent sample t-test and Fast Fourier Transform (FFT). The CNN-SE Attention-LSTM (CAL) model, with hyperparameters optimized by the Osprey Optimization Algorithm (OOA), serves as the primary forecasting engine. In addition, a dedicated heterogeneous error correction module, namely TTL, is introduced, in which Temporal Convolutional Network, Transformer, and LSTM are designed to capture local transients, long-range dependencies, and transitional dynamics in the residual series, respectively. Empirical results demonstrate that compared with the Naive benchmark, COCAL-TTL achieves percentage MAPE improvements of 58.48% and 48.97% in low- and high-volatility regimes, respectively. These findings indicate that the proposed structural break-aware framework provides a robust data-driven solution for EPF under heterogeneous market conditions and offers technical support for stable electricity market operation in the context of renewable energy integration. Full article

To address the inherent limitations of the Dhole Optimization Algorithm (DOA)—limited exploration range, insufficient population diversity, and slow convergence—this paper proposes a Modified Dhole Optimization Algorithm (MDOA) integrating a Beta distribution-based opposition learning strategy, a DE/rand-to-best/1 differential mutation mechanism, and nonlinear parameter control. MDOA is evaluated on 41 CEC2017 and CEC2022 benchmark functions, outperforming 11 state-of-the-art algorithms in convergence speed, accuracy, and robustness. It is then applied to five engineering optimization problems: compression spring design, speed reducer weight minimization, rolling bearing optimization, tubular column design, and moisture content prediction of Dendrobium huoshanense using near-infrared spectroscopy with a BP neural network. The MDOA-BP model reduces MAE, RMSE, MSE, and MAPE by 27.5%, 27.8%, 47.6%, and 31.0%, respectively, while increasing $R^2$ from 0.8339 to 0.9130, achieving the best results among all comparison models. These results demonstrate that MDOA is a highly effective and robust optimizer for complex constrained engineering and high-dimensional optimization tasks. Full article

Background/Objectives: In gas chromatography–mass spectrometry (GC-MS) library-based compound identification, spectrum preprocessing and associated tuning parameters critically influence identification performance. These parameters are conventionally optimized using grid search, which requires predefined parameter spaces and becomes computationally inefficient as dimensionality increases, often failing to identify optimal values because of discretization. Differential evolution (DE), a population-based metaheuristic optimization algorithm, provides a flexible alternative through efficient global exploration of the parameter space. This study compared the performance of DE and grid search for optimizing compound identification. Methods: Cosine similarity was applied to the NIST GC-MS library. DE was used to maximize either cross-validated accuracy or mean reciprocal rank (MRR). Results were compared with those from a grid search over five equally spaced parameter values. Identification performance was evaluated using accuracy, MRR, and area under the receiver operating characteristic curve (AUC). Results: When all four parameters were optimized simultaneously, DE achieved slightly higher cross-validated accuracy and MRR than grid search, although the absolute differences were modest. More pronounced differences were observed in specific unidimensional tuning scenarios, particularly for the intensity weight factor. Simultaneous multidimensional parameter optimization yielded better performance than isolated parameter tuning. Conclusions: Grid search may be computationally advantageous when the parameter space is known and limited, whereas DE provides a more flexible approach for unknown or high-dimensional search spaces. Overall, DE achieved comparable identification performance to grid search, with modest improvements observed in some optimization settings. A command line Julia-based tool, MSTune, was developed for spectrum preprocessing parameter optimization and is publicly available on GitHub. Full article

Predictive maintenance in underground mining faces challenges due to severe conditions such as confined environments, high humidity, presence of silica dust, and restricted access. This study develops a predictive framework based on oil analysis and machine learning for multiple compartments of mining equipment (engine, hydraulic system, transmission, differential). Samples were processed under ASTM standards, integrating wear metal concentrations (Fe, Cu, Cr, Pb, Al), physicochemical properties (viscosity, TBN, soot), and contaminants (Si, Na). Based on tribology, interpretable ratios were constructed. Three algorithms (Random Forest, Gradient Boosting, and XGBoost) were evaluated using cross-validation. XGBoost achieved the best balance (F1 = 0.852, AUC = 0.975), with a recall of 94.5% for the critical class and only 3 false negatives out of 199 test samples, while Random Forest presented the highest global discrimination power (AUC = 0.978). SHAP revealed that viscosity at 100 °C is the most important predictor (SHAP ~0.9), surpassing iron. No temporal wear trend was found (R² = 0.000). Threshold optimization to 0.25 reduced false negatives by 67% (from 9 to 3). The framework provides interpretable predictions with uncertainty quantification for underground environments. Full article