Search Results (1,363)

Three-dimensional path planning is critical for the successful operation of unmanned aerial vehicles (UAVs), automated guided vehicles (AGVs), and robots in industrial Internet of Things (IIoT) applications. In 3D path planning, the standard Particle Swarm Optimization (PSO) algorithm suffers from premature convergence and a tendency to fall into local optima, leading to significant deviations from the optimal path. This paper proposes an improved PSO (IPSO) algorithm that enhances particle diversity and randomness through the introduction of logistic chaotic mapping, while employing dynamic learning factors and nonlinear inertia weights to improve global search capability. Experimental results demonstrate that IPSO outperforms traditional methods in terms of path length and computational efficiency, showing potential for real-time path planning in complex environments. Full article

The accuracy of modal superposition methods for determining displacement or strain field of structures largely depends on the selection of modes relevant to its deformation. Analytical methods for modal selection have been developed to minimise errors in reconstructing deformation through a linear combination of modal shapes. This study constitutes an initial step towards the development of structural health-monitoring algorithms for large engineering machines, where continuous monitoring of strain and stress, assuming a linear elastic field, is critical. The focus is on selecting modes that significantly contribute to the reconstruction of static deformation of structures. A detailed analytical approach, derived from established structural dynamics principles, leads to the formulation of modal selection criteria. These criteria are based on two fundamental quantities from dynamic and elastic theory: the modal participation factor and internal strain potential energy. Three criteria are introduced: the directional participation factor criterion (DPFC), the global participation factor criterion (GPFC), and the internal strain potential energy criterion (ISPEC). While DPFC and GPFC rely on displacements, ISPEC uses strains. The methods are validated through a case study involving a rectangular plate subjected to various loads, demonstrating their applicability to complex deformation scenarios, which require the combination of multiple modes to fully describe the static deformation. The proposed criteria are formulated for linear elastic systems and are therefore applicable, in principle, to plate-like components, machine casings, thin structural panels, and certain civil and aerospace panels, under the assumptions of small strains, linear constitutive behaviour, and validity of modal superposition. The approach also represents a first step towards the integration of modal selection with machine learning for structural health-monitoring applications and presents a computational cost significantly lower than that of full finite element analyses. Full article

As critical infrastructure for flood control and disaster mitigation, the completeness of a dam spatial database directly impacts regional emergency disaster response. However, existing dam data in some developing countries suffer from severe gaps and outdated information, particularly concerning small- and medium-sized dams, hindering rapid response during disasters. There is an urgent need to improve the physical dam database and implement dynamic monitoring. Yet, current remote sensing identification methods face limitations, including a lack of diverse dam samples, limited analysis of geographical factors, and low efficiency in full-image processing, making it difficult to efficiently enhance dam databases. To address these issues, this study proposes a dam extraction framework integrating comprehensive geographical factor analysis with deep learning detection, validated in Sindh Province, Pakistan. Firstly, multiple geographical factors were fused using the Random Forest algorithm to generate a dam existence probability map. High-probability candidate areas were delineated using dynamic threshold segmentation (precision: 0.90, recall: 0.76, AUC: 0.86). Subsequently, OpenStreetMap (OSM) water body data excluded non-dam potential areas, further narrowing the candidate areas. Finally, a dam image dataset was constructed to train a dam identification model based on YOLOv11, achieving an mAP50 of 0.85. This trained model was then applied to high-resolution remote sensing imagery of the candidate areas for precise identification. Ultimately, 16 previously unrecorded small and medium-sized dams were identified in Sindh Province, enhancing its dam location database. Experiments demonstrate that this method, through the synergistic optimization of geographical constraints and deep learning, significantly improves the efficiency and reliability of dam identification. It provides high-precision data support for dam disaster emergency response and water resource management, exhibiting strong practical utility and regional scalability. Full article

Keratinocytes are the primary cells of the epidermis layer in our skin. They play a crucial role in maintaining skin health, responding to injuries, and counteracting disease progression. Understanding their behaviour is essential for advancing wound healing therapies, improving outcomes in regenerative medicine, and developing numerical models that accurately mimic skin deformation. To create physically representative models, it is essential to evaluate the nuanced ways in which keratinocytes deform, interact, and respond to mechanical and biochemical signals. This has prompted researchers to investigate various computational methods that capture these dynamics effectively. This review summarises the main mathematical and biomechanical modelling techniques (with particular focus on the literature published since 2010). It includes reaction–diffusion frameworks, finite element analysis, viscoelastic models, stochastic simulations, and agent-based approaches. We also highlight how machine learning is being integrated to accelerate model calibration, improve image-based analyses, and enhance predictive simulations. While these models have significantly improved our understanding of keratinocyte function, many approaches rely on idealised assumptions. These may be two-dimensional unicellular analysis, simplistic material properties, or uncoupled analyses between mechanical and biochemical factors. We discuss the need for multiscale, integrative modelling frameworks that bridge these computational and experimental approaches. A more holistic representation of keratinocyte behaviour could enhance the development of personalised therapies, improve disease modelling, and refine bioengineered skin substitutes for clinical applications. Full article

Decentralized electricity markets are increasingly shaped by the proliferation of distributed energy resources, the rise of prosumers, and growing demands for privacy-aware analytics. In this context, federated learning (FL) emerges as a promising paradigm that enables collaborative model training without centralized data aggregation. This review systematically explores the application of FL in energy systems, with particular attention to architectures, heterogeneity management, optimization tasks, and real-world use cases such as load forecasting, market bidding, congestion control, and predictive maintenance. The article critically examines evaluation practices, reproducibility issues, regulatory ambiguities, ethical implications, and interoperability barriers. It highlights the limitations of current benchmarking approaches and calls for domain-specific FL simulation environments. By mapping the intersection of technical design, market dynamics, and institutional constraints, the article formulates a pluralistic research agenda for scalable, fair, and secure FL deployments in modern electricity systems. This work positions FL not merely as a technical innovation but as a socio-technical intervention, requiring co-design across engineering, policy, and human factors. Full article

This paper proposes an adaptive path-planning algorithm for unmanned ground vehicles (UGVs) in three-dimensional terrain environments. The algorithm first constructs an interference model between the UGV chassis and the three-dimensional terrain, taking into account the impact of terrain undulations on vehicle driving stability. A dynamic cost-adjustment mechanism for multi-task modes was designed, which introduces a learning model to automatically identify task types and dynamically adjust the weights of various cost factors in path planning accordingly. This paper constructs simulation environments for sparse obstacle scenes and high-density obstacle scenes, respectively, to verify the effectiveness of the path-planning results of the algorithm in different task modes. The experimental results show that the proposed method can generate smoother, safer, and more task-matched trajectory paths while ensuring path feasibility, verifying the good adaptability and robustness of this algorithm for complex unstructured environments under multi-task driving conditions. Full article

Regional drought monitoring based on the Temperature Vegetation Drought Index (TVDI) holds significant potential in efforts to ensure food safety. However, its empirical determination of dry and wet edges introduces subjectivity and uncertainty, limiting its accuracy and applicability. An improved TVDI (iTVDI) was developed by optimizing boundary parameters using reinforcement learning, based on maximizing the correlation between the TVDI and the ERA5-Land soil moisture dataset. The findings are as follows: (1) The enclosed area and the absolute value of dry edge slope of iTVDI was 34.83–39.97% and 0.79–33.75% larger than TVDI, indicating that the iTVDI can be used to achieve better representation of drought conditions. (2) The iTVDI showed stronger correlations with ERA5 soil moisture (r: −0.416 to −0.174), with average |r| values 17.25% higher than TVDI; its correlations with Standardized Precipitation Index (SPI), the Standardized Precipitation Evapotranspiration Index (SPEI), and the Vegetation Condition Index (VCI) were also 12.69–75.43% higher. (3) From 2005 to 2024, the spring drought in the Huaihe Basin intensified, with the annual iTVDI increasing by 0.008–0.011, primarily driven by rising temperature, potential evapotranspiration, and vapor pressure deficit. Overall, the iTVDI is proved to be more accurate and reliable for monitoring drought dynamics and driving factors. Full article

Under varying operating conditions, motor bearings undergo continuous changes, necessitating the development of deep learning models capable of robust fault diagnosis. While meta-learning can enhance generalization in low-data scenarios, it is often susceptible to overfitting. Domain adaptation mitigates this by aligning feature distributions across domains; however, most existing methods primarily focus on global alignment, overlooking intra-class subdomain variations. To address these limitations, we propose a novel Dynamic Balance Domain-Adaptation based Few-shot Diagnosis framework (DBDA-FD), which incorporates both global and subdomain alignment mechanisms along with a dynamic balancing factor that adaptively adjusts their relative contributions during training. Furthermore, the proposed framework implicitly leverages the concept of symmetry in feature distributions. By simultaneously aligning global and subdomain-level representations, DBDA-FD enforces a symmetric structure between source and target domains, which enhances generalization and stability under varying operational conditions. Extensive experiments on the CWRU and PU datasets demonstrate the effectiveness of DBDA-FD, achieving 97.6% and 97.3% accuracy on five-way five-shot and three-way five-shot tasks, respectively. Compared to state-of-the-art baselines such as PMML and ADMTL, our method achieves up to 1.4% improvement in accuracy while also exhibiting enhanced robustness against domain shifts and class imbalance. Full article

Knee osteoarthritis (KOA) is a most prevalent chronic muscoloskeletal disorder causing pain and functional impairment. Accurate predictions of KOA evolution are important for early interventions and preventive treatment planning. In this paper, we propose a novel dynamic hypergraph-based risk model (DyHRM) which integrates the encoder–decoder (ED) architecture with hypergraph convolutional neural networks (HGCNs). The risk model is used to generate longitudinal forecasts of KOA incidence and progression based on the knee evolution at a historical stage. DyHRM comprises two main parts, namely the dynamic hypergraph gated recurrent unit (DyHGRU) and the multi-view HGCN (MHGCN) networks. The ED-based DyHGRU follows the sequence-to-sequence learning approach. The encoder first transforms a knee sequence at the historical stage into a sequence of hidden states in a latent space. The Attention-based Context Transformer (ACT) is designed to identify important temporal trends in the encoder’s state sequence, while the decoder is used to generate sequences of KOA progression, at the prediction stage. MHGCN conducts multi-view spatial HGCN convolutions of the original knee data at each step of the historic stage. The aim is to acquire more comprehensive feature representations of nodes by exploiting different hyperedges (views), including the global shape descriptors of the cartilage volume, the injury history, and the demographic risk factors. In addition to DyHRM, we also propose the HyGraphSMOTE method to confront the inherent class imbalance problem in KOA datasets, between the knee progressors (minority) and non-progressors (majority). Embedded in MHGCN, the HyGraphSMOTE algorithm tackles data balancing in a systematic way, by generating new synthetic node sequences of the minority class via interpolation. Extensive experiments are conducted using the Osteoarthritis Initiative (OAI) cohort to validate the accuracy of longitudinal predictions acquired by DyHRM under different definition criteria of KOA incidence and progression. The basic finding of the experiments is that the larger the historic depth, the higher the accuracy of the obtained forecasts ahead. Comparative results demonstrate the efficacy of DyHRM against other state-of-the-art methods in this field. Full article

Objective: The aim of this study was to evaluate the accuracy of machine and deep learning approaches on radiomics features obtained by Dynamic Contrast Enhanced Magnetic Resonance Imaging (DCE-MRI) and contrast enhanced mammography (CEM) in the characterization of breast cancer and in the prediction of the tumor molecular profile. Methods: A total of 153 patients with malignant and benign lesions were analyzed and underwent MRI examinations. Considering the histological findings as the ground truth, three different types of findings were used in the analysis: (1) benign versus malignant lesions; (2) G1 + G2 vs. G3 classification; (3) the presence of human epidermal growth factor receptor 2 (HER2+ vs. HER2−). Radiomic features (n = 851) were extracted from manually segmented regions of interest using the PyRadiomics platform, following IBSI-compliant protocols. Highly correlated features were excluded, and the remaining features were standardized using z-score normalization. A feature selection process based on Elastic Net regularization (α = 0.5) was used to reduce dimensionality. Synthetic balancing of the training data was applied using the ROSE method to address class imbalance. Model performance was evaluated using repeated 10-fold cross-validation and AUC-based metrics. Results: Among the 153 patients enrolled in the studies, 113 were malignant lesions. Among the 113 malignant lesions, 32 had high grading (G3) and 66 had the HER2+ receptor. Radiomic features derived from both CEM and DCE-MRI showed strong discriminative performance for malignancy detection, with several features achieving AUCs above 0.80. Gradient Boosting Machine (GBM) achieved the highest accuracy (0.911) and AUC (0.907) in differentiating benign from malignant lesions. For tumor grading, the neural network model attained the best accuracy (0.848), while LASSO yielded the highest sensitivity (0.667) for detecting high-grade tumors. In predicting HER2+ status, the neural network also performed best (AUC = 0.669), with a sensitivity of 0.842. Conclusions: Radiomics-based machine learning models applied to multiparametric CEM and DCE-MRI images offer promising, non-invasive tools for breast cancer characterization. The models effectively distinguished benign from malignant lesions and showed potential in predicting histological grade and HER2 status. These results demonstrate that radiomic features extracted from CEM and DCE-MRI, when analyzed through machine and deep learning models, can support accurate breast cancer characterization. Such models may assist clinicians in early diagnosis, histological grading, and biomarker assessment, potentially enhancing personalized treatment planning and non-invasive decision-making in routine practice. Full article

Human energy production and consumption have significantly contributed to the environmental crisis, impacting human health, wellbeing, and social justice. In this context, the concept of energy citizenship has emerged, referring to civic engagement in fostering sustainable and democratic energy systems and transitions. Under the Horizon Europe project EnergyPROSPECTS (PROactive Strategies and Policies for Energy Citizenship Transformation), we investigated the conditions and dynamics that promote or hinder energy citizenship and empower citizens to contribute to sustainable energy transformations. Through 44 in-depth interviews and four deliberative workshops in four European case study regions with individuals and organizations engaged in different forms of energy citizenship, we identified key psychological and organizational factors driving citizen empowerment. These findings informed the development of an interactive empowerment toolkit, a digital learning resource designed to enhance energy citizenship literacy and skills. This toolkit, although primarily targeting adults interested in energy citizenship, is adaptable for students and educators at various levels, offering two tracks: one for beginners with no prior involvement in the exercise of energy citizenship, and another for those with experience in energy activism. We highlight the scientific basis of the toolkit, detailing its components and demonstrating its application in fostering energy citizenship empowerment. The tool aims to equip users with the skills and knowledge necessary to actively participate in sustainable energy transitions. Full article

The black soil region of northeast China is a critical global grain production base. The dynamic variations in soil organic carbon (SOC) are directly linked to the regional food security. To accurately monitor SOC content and evaluate the potential of integrating Landsat-9 and GF-1 satellite data for SOC inversion, we developed a machine learning framework that combines data from both satellite sources to model SOC. Using the typical black soil region of northeast China in the Tongken River Basin as the study area, we compared the MLR, PLSR, RF, and XGBoost algorithms. And XGBoost demonstrated the highest performance (R² = 0.9130; RMSE = 0.3834%). Based on the optimal model, SOC in the study area was projected from 2020 to 2024. The multi-year average SOC exhibited an initial increase followed by a subsequent decline, with an overall increase of 22.78%. Spearman correlation analysis identified parent material as the dominant factor controlling SOC variation at the watershed scale (correlation coefficient = 0.38) while also modulating the influence of land use types on SOC dynamics. The “space–ground” multi-source collaborative inversion framework developed in this study offers a high-precision technical approach for the monitoring of SOC in black soil regions. Full article

Robots play a crucial role in experimental smart cities and are ubiquitous in daily life, especially in complex environments where multiple robots are often needed to solve problems collaboratively. Researchers have found that the swarm intelligence optimization algorithm has a better performance in planning robot paths, but the traditional swarm intelligence algorithm cannot be targeted to solve the robot path planning problem in difficult problem. Therefore, this paper aims to introduce a fuzzy controller, mutation factor, exponential noise, and other strategies on the basis of particle swarm optimization to solve this problem. By judging the moving speed of different particles at different periods of the algorithm, the individual learning factor and social learning factor of the particles are obtained by fuzzy controller, and using the leader particle and random particle, designing a new dynamic balance of mutation factor, with the iterative process of the adaptation value of continuous non-updating counter and continuous updating counter to control the proportion of the elite individuals and random individuals. Finally, using exponential noise to update the matrix of the population every 50 iterations is a way to balance the local search ability and global exploration ability of the algorithm. In order to test the proposed algorithm, the main method of this paper is simulated on simple scenarios, complex scenarios, and random maps consisting of different numbers of static obstacles and dynamic obstacles, and the algorithm proposed in this paper is compared with eight other algorithms. The average path deviation error of the planned paths is smaller; the average distance of untraveled target is shorter; the number of steps of the robot movements is smaller, and the path is shorter, which is superior to the other eight algorithms. This superiority in solving multi-robot cooperative path planning has good practicality in many fields such as logistics and distribution, industrial automation operation, and so on. Full article

Oral tongue squamous cell carcinoma (OTSCC) is a common type of oral cancer influenced by genetic, epigenetic, and environmental factors like exposure to environmental toxins. These environmental toxins can decrease the effectiveness of established chemotherapy drugs, such as Irinotecan, used in OTSCC treatment. Bioinformatics, drug discovery, and machine learning techniques were employed to investigate the impact of Irinotecan on OTSCC patients by identifying targets and signaling pathways, including those that positively influence protein phosphorylation, protein tyrosine kinase activity, the PI3K-Akt (Phosphatidylinositol 3-kinase- Protein Kinase B) signaling system, cancer pathways, focal adhesion, and the HIF-1 (Hypoxia-Inducible Factor 1) signaling pathway. Later, the protein–protein interactions (PPIs) network, along with twelve cytoHubba approaches to finding the most interacting molecule, was employed to find the important proteins BCL2 and EGFR. Drugs related to BCL2 and EGFR were extracted from the DGIdb database for further molecular docking. Molecular docking revealed that Docetaxel, Paclitaxel, Imatinib, Ponatinib, Ibrutinib, Sorafenib, and Etoposide showed more binding affinity than Irinotecan (i.e., −9.8, −9.6). Of these, Paclitaxel (−10.3, −11.4) and Imatinib (−9.9, −10.4) are common in targeting BCL2 and EGFR. Using these identified candidate genes and pathways, we may be able to uncover new therapeutic targets for the treatment of OTSCC. Furthermore, molecular dynamics (MD) simulations were performed for selected ligand–receptor complexes, revealing stable binding interactions and favorable energetic profiles that supported the docking results and strengthened the reliability of the proposed drug repurposing strategy. Full article

Limited evidence exists on how teachers contribute to student learning gains in early childhood education. This study draws on the Dynamic Model of Educational Effectiveness (DMEE) and investigates the impact of teacher factors on pre-primary students’ mathematics achievement. It also examines whether the five proposed dimensions—frequency, quality, focus, stage, and differentiation—can clarify the conditions under which these factors influence learning. Using a stage sampling procedure, 463 students and 27 teachers from Greek pre-primary schools were selected. Mathematics achievement was assessed at the beginning and end of the school year, while external observations measured the DMEE factors. Analysis of observation data using multi-trait multilevel models provided support for the construct validity of the measurement framework. Teacher factors explained variation in student achievement gains in mathematics. The added value of using a multidimensional approach to measure the functioning of the teacher factor was identified. Implications of the findings are drawn. Full article