Search Results (2,034)

Background/Objectives: Hyperreflective foci (HRF) are supportive optical coherence tomography (OCT) imaging biomarkers that have been examined for their association with disease progression and severity in various retinal disorders. The accurate identification and segmentation of these tiny structures of lipid extravasation remain complicated because of their small size, class imbalance, similarity in the reflectivity patterns with the surrounding structures and imaging artifacts. While U-Net-based models have promised exceptional results for medical image segmentation, optimal architectural settings and suitable preprocessing methods for HRF detection remain unclear. Methods: This research assessed optimal settings for U-Net-based models for HRF segmentation by evaluating standard U-Net and attention U-Net under different preprocessing regimes. Attention U-Net employed Z-score normalization and contrast-limited adaptive histogram equalization (CLAHE) enhancement with soft dice loss. The standard U-Net was trained on OCT images with CLAHE using focal Tversky loss. A total of 435 fovea-centered OCT B scans with the corresponding, consensus-annotated HRF masks were utilized for this research. Results: The standard U-Net outperformed attention U-Net with a dice score of 0.5207, an AUC of 0.8411, and a recall of 0.6439 on raw OCT images. The attention U-Net with preprocessing (dice: 0.5033, AUC: 0.6987, recall: 0.5391) demonstrated satisfactory performance. The results showed that the U-Net model with CLAHE and focal Tversky loss improved recall by 19.4% relative to the attention U-Net, and this corresponds roughly to a 23% relative decline in false negatives. This indicates increased sensitivity in identifying HRF regions. Conclusions: The best-performing configuration using U-Net-based architectures for segmentation of HRFs combines the standard U-Net model with CLAHE and focal Tversky loss for handling class imbalance. This approach yields relatively higher sensitivity, indicating that the standard U-Net model delivers a simple and robust framework for automated HRF segmentation on the evaluated dataset, promising further validation in broader clinical datasets. Full article

In large-scale dam construction, the efficiency of concrete transport operations is fundamentally governed by the coordination between horizontal hauling and vertical hoisting capacities. Traditional experience-based scheduling approaches often fail to capture the stochastic, cyclic, and resource-coupled nature of these transport systems. This study developed a closed queuing network-based stochastic simulation framework to model dam concrete transportation as a finite-population cyclic service system. The process was abstracted into sequential service stages with stochastic service times, and a structured state-space representation combined with time-step simulation was constructed to describe dynamic resource occupation and task transitions under varying truck and cable crane configurations. Application to a real large-scale dam project revealed a characteristic multi-stage performance evolution pattern governed by capacity matching mechanisms. As the truck fleet size increased, system performance transitioned from a transport-limited regime to a capacity-coordination regime and ultimately to a hoisting-saturated regime in which further fleet expansion yielded diminishing returns. Sensitivity analysis demonstrated that hoisting capacity imposed an upper bound on system throughput, while adaptive fleet reconfiguration could restore operational equilibrium under constrained equipment availability. The results indicated that dam concrete transport should be treated as a dynamic capacity regulation problem rather than a static allocation task. The proposed framework provides an interpretable and quantitative decision-support tool for equipment configuration, bottleneck identification, and adaptive scheduling in large-scale hydraulic infrastructure projects. Full article

Accurate prediction of contact temperature in disconnectors is critical for early fault detection. However, purely physics-based models face difficulties in parameter identification, while purely data-driven models often suffer from error accumulation in long-term forecasting. To address these challenges, this paper proposes a novel framework named Hybrid Physics-Informed Long Short-Term Memory (Hybrid-PI-LSTM). Firstly, this paper mathematically formulates the transient heat transfer process as a constrained optimization problem governed by a nonlinear ordinary differential equation (ODE), embedding physical laws into the loss function as a regularization term to promote dynamic consistency. Secondly, to address the inverse problem of parameter drift caused by environmental changes, an Adaptive Parameter Updating (APU) mechanism is introduced. This algorithm utilizes a gradient-based iterative approach to dynamically estimate equivalent physical coefficients (e.g., heat capacity) from observational residuals during inference. Finally, numerical experiments on a real-world dataset demonstrate that the proposed framework significantly outperforms baseline models. Specifically, it achieves a Root Mean Squared Error (RMSE) of 0.283 at a 720-step forecasting horizon, reducing the prediction error by over 35% compared to static-parameter physical models. The results indicate that the proposed adaptive constraint mechanism contributes to enhanced long-term numerical stability and physics-guided parameter tracking. Full article

To address the challenges commonly encountered in civil aircraft operating under multi-mode, strongly coupled closed-loop control—namely scarce fault samples, pronounced distribution shift, and root-cause explanations that are easily confounded by covariates—this paper proposes a root-cause analysis method that integrates ensemble learning with constraint-guided reasoning by large language models (LLMs). First, for Full Authority Digital Engine Control (FADEC) monitoring sequences, a feature system comprising environment-normalized ratios, mechanism-informed mixing indices, and multi-scale temporal statistics is constructed, thereby improving cross-mode comparability and enhancing engineering-semantic expressiveness. Second, in the anomaly detection stage, a cost-sensitive LightGBM model is adopted and a validation-set-based adaptive thresholding strategy is introduced to achieve robust identification under highly imbalanced fault conditions. Furthermore, for Root Cause Analysis (RCA), a “computation–reasoning decoupling” framework is developed: Shapley Additive exPlanations (SHAP) are used to generate segment-level contribution evidence, while causal chains, engineering prohibitions, and structured output templates are injected into prompts to constrain the LLM, enabling it to infer root-cause candidates and produce structured explanations under mechanism-consistency constraints. Experiments on real flight data demonstrate that our method yields an anomaly detection F1-score of 0.9577 and improves overall RCA accuracy to 97.1% (versus 62.3% for a pure SHAP baseline). Practically, by translating complex high-dimensional data into actionable natural language diagnostic reports, the proposed method provides reliable and interpretable decision support for rapid RCA. Full article

Aerobic composting shows state-dependent dynamics in key parameters such as temperature, moisture content, oxygen concentration, and pH, and these variables are strongly coupled over time. This coupling makes accurate state identification and process regulation challenging when relying on single-parameter thresholds or experience-based control. To enable robust recognition of composting states throughout the process, we propose an IRRTO-optimized CNN–LSTM–attention model (IRRTO–CNN–LSTM–attention). The model uses a convolutional neural network (CNN) to extract discriminative multivariate features, a long short-term memory (LSTM) network to model temporal dependencies, and an attention module to adaptively emphasize informative features. To address the hyperparameter selection challenge, the Rapidly-exploring Random Tree Optimizer (RRTO) was introduced and further enhanced via four strategies (fluctuating attenuation adaptive regulation, dual-mode guided update, dynamic dimension adaptive perturbation, and dual-mechanism adaptive perturbation regulation), forming the improved IRRTO. The proposed approach was validated using sensor data from windrow composting of pig manure and corn straw. The IRRTO–CNN–LSTM–attention model achieved an overall accuracy of 98.31% in classifying the four states (mesophilic/heating, thermophilic, cooling, and abnormal) on the independent test set, which was 3.39 percentage points higher than the RRTO-based model. These results suggest that the proposed method can accurately identify composting states and support early warning and state-specific regulation in practical aerobic composting systems. Full article

Scaling up microalgae cultivation is key to commercial viability. Over the past two decades, the market value of microalgae has expanded exponentially, driven by their applications in the pharmaceutical, nutraceutical, cosmetic, and animal feed industries. High-value compounds such as omega-3 fatty acids, proteins, and pigments are in strong demand. However, supply remains constrained by suboptimal cultivation practices and high harvesting costs. Despite decades of progress in process modeling, control, and optimization, industrial adoption is still limited by dynamic cultivation conditions influenced by weather variability, biological adaptation, and integration challenges. Technical barriers, including limited data accuracy, modest control performance, and the fragility of low-cost devices, further restrict optimization efforts. In response, we examined recent advances in control, optimization, and automated machine learning applied to microalgae cultivation. We propose an automated architecture built on a closed-loop supervisory layer that embeds machine learning within the control loop, enabling real-time monitoring, prediction, and adaptive actuation. This approach aligns with real-time optimization and distributed control system practices, integrating system identification, controller optimization, fault diagnosis and tolerance, and perception to achieve autonomous, uncertainty-aware operation. Full article

Background: Emotion regulation refers to the processes through which individuals influence their emotional experiences, including how emotions are generated, experienced, and expressed. Difficulties in emotion regulation have been identified as a relevant factor in the development and maintenance of Eating Disorders (EDs). In the sports context, high physical and performance demands may intensify emotional challenges, potentially increasing vulnerability to eating disorder symptomatology among athletes. Objectives: This systematic review aimed to examine the relationship between emotion regulation and EDs in athletic populations, with a particular focus on emotion regulation strategies and related emotional processes. Methods: The PICO model was used, and PRISMA guidelines were followed. The Redalyc, Dialnet, SpringerLink, and PubMed databases were searched from inception to April 2025, with an update in November 2025. After the selection process, nine studies involving athletes from different disciplines and competitive levels were included. Methodological quality and risk of bias were assessed using the Joanna Briggs Institute (JBI) Critical Appraisal Checklists. Results: The findings indicate that adaptive emotion regulation strategies, such as Cognitive Reappraisal and emotional identification, are associated with lower levels of eating disorder symptomatology, body dissatisfaction, and greater resilience to sport-related pressures. In contrast, dysfunctional strategies, including expressive suppression, emotional unawareness, and difficulties in emotion management, were consistently associated with restrictive eating behaviors, bulimic symptomatology, excessive weight control, and increased ED risk. Additional emotional factors, including anxiety, perfectionism, low self-esteem, and body image dissatisfaction, were also related to higher vulnerability to EDs, particularly in sports with high aesthetic or weight-related demands. Conclusions: Emotional regulation is closely associated with ED risk in athletes. Adaptive emotion regulation strategies may serve as protective factors, whereas dysfunctional strategies are associated with increased risk. Full article

Background/Objectives: Children and adolescents who care for family members with illness, disability, or mental health conditions face challenges across educational, health, and psychosocial domains. Although research and practice have developed conceptual models and assessment tools to better understand and address young carers’ situations, a persistent gap remains between their needs and available support, reflecting structural fragmentation across health, education, and social care systems. To address this gap, this article presents the development of a family-centred framework spanning these sectors. Methods: The framework was developed through an iterative, empirically grounded process based on two studies within a larger research project on young carers in Switzerland. Key themes, structural challenges, and support-related factors were identified by systematically synthesising the findings of the two studies and integrated into an overarching framework linking young carers’ family contexts with cross-sectoral service structures. Results: The Caring in Context Framework synthesises empirical findings into a coherent framework for understanding and addressing young carers’ situations. By systematically extending the whole family approach to include a cross-sectoral dimension, it bridges relational family dynamics and structural service contexts. Sustainable support is conceptualised as dependent on the structural visibility and institutional recognition of young carers across all system levels, positioning identification and recognition as prerequisites for coordinated responses in research, policy, and practice. Conclusion: The framework advances conceptual clarity by integrating family-centred and cross-sectoral perspectives. Rather than creating new services, it emphasises adapting and coordinating existing structures while ensuring systematic recognition of young carers to support coherent, sustainable, and inclusive strategies. Full article

Identifying the time-domain waveform type under broadband conditions is a basic but very challenging task. Traditional methods based on frequency domain or training models generally have the problems of high resource consumption, large delay, and unsuitability for hardware. This paper proposes a time-domain waveform recognition architecture based on an FPGA, which is integrated with multi-feature voting. Several lightweight time domain characteristics, such as high amplitude ratio, symmetry, slope uniformity, slope change rate, and flat-top characteristics, are extracted and directly used for waveform classification. Then classify sine waves, square waves, triangular waves, and noise in the time domain according to the decision-making mechanism of voting. In order to improve reliability under non-ideal conditions, adaptive thresholds and noise perception decision-making logic are used to suppress misclassifications caused by random fluctuations and jitter. The whole engineering design focuses on resource consumption and hardware efficiency, using a fully pipeline FPGA architecture. The experimental results prove that the system has the ability of high-precision identification, low power consumption, and real-time processing in the wide frequency band, providing an efficient and practical solution for embedded waveform recognition applications. Full article

The adoption of digital twins has revolutionized industrial process simulation, monitoring, and control effectiveness. However, practical implementations of digital twins are hindered by substantial challenges, including extended development time, diminishing model accuracy, and restricted interactive capabilities. Addressing these critical issues, this paper proposes a comprehensive digital twin development framework that integrates digital twin identification, real-time model updating, and advanced process control. The proposed approach first identifies the offline digital twin model through the sparse identification of a nonlinear dynamics algorithm, reducing the digital twin development time while maintaining model fidelity. Then, the identified model is updated by the extended Kalman filter to mitigate the problem of diminishing accuracy. Finally, incorporating the latest updated model into the model predictive control facilitates the control inputs optimization and enhances the interactive capacity of digital twins. Through one industrial case study and two simulation examples, the advantages of the proposed algorithm are demonstrated. Full article

Radio frequency fingerprints (RFFs), arising from inherent hardware imperfections, serve as distinctive features for device identification. The location- and time-dependent nature of the wireless channel directly affects RFF-based device identification, making it challenging under different channel conditions. This is primarily because the training and test datasets containing RFFs may not overlap within the same feature-space domain. In this work, the mentioned issue is addressed as a domain adaptation problem. For this objective, we propose the use of a triplet-learning-based domain-adversarial neural network within a hybrid framework named TripletDANN. We leverage the triplet loss, enabling the network to focus exclusively on device-specific latent representations under different channel conditions, while employing an adversarial loss to prevent the network from exploiting channel-specific characteristics. With this aim, data aggregation is performed together with channel labeling. The generalization capability of TripletDANN is evaluated on previously unseen test data collected across different locations under two distinct scenarios. Raw I/Q signals of 15 Wi-Fi devices are used as a case study. The proposed TripletDANN model achieves up to $88.52\%$ average device classification accuracy across the different data collection locations. On average, TripletDANN attains up to a $5\%$ performance improvement over its counterpart model. Moreover, data augmentation is employed to improve the overall performance, and a highest accuracy of $96.71\%$ is achieved on experimentally collected test data from an unseen location. Full article

This paper presents the design and industrial implementation of an adaptive discrete control system for a rotary dryer operating in potash fertilizer production. The drying process is characterized by high inertia, multivariable interactions, transport delay, and non-stationary behavior resulting from variations in raw material properties and external disturbances, which significantly reduce the effectiveness of conventional fixed-parameter controllers. A discrete-time mathematical model of the rotary drying process was developed using industrial experimental data collected from a full-scale production plant. The process was modeled as a coupled 2 × 2 multivariable system with pronounced time-delay effects in the main control channels. System identification was carried out using statistical and frequency-domain methods to capture the dominant dynamic characteristics required for controller synthesis. Based on the identified model, an adaptive discrete controller with online parameter adjustment was developed to regulate outlet moisture content and exhaust gas temperature. Simulation and industrial results confirmed stable operation under varying conditions, improved regulation accuracy, enhanced process stability, and an average production efficiency increase of approximately 1.8%, accompanied by reduced fuel consumption. Full article

Background/Objectives: Accurate brain tumor segmentation from MRI is crucial for diagnosis but faces challenges like domain shifts across medical centers, data privacy constraints, and high annotation costs. While source-free active domain adaptation (SFADA) emerges as a promising solution to these issues, existing approaches often overlook the inherent structural complexity in tumor regions. Methods: We propose a novel SFADA framework composed of two major contributions. First, we introduce a Region-level Uncertainty-Guided Sample Selection (RUGS) strategy, enabling the identification of the most informative target-domain samples in a single inference pass. Second, we present the Source-Free Active Domain Adaptation Network (SFADA-Net), a Mamba-driven segmentation model equipped with a dual-path multi-kernel convolution module for enhanced local feature interaction and a structure-aware prompted Mamba module for capturing global spatial relationships. Results: Extensive evaluations across one source domain dataset (BraTS-2021) and three target domain datasets (BraTS-SSA, BraTS-PED, and BraTS-MEN 2023) demonstrate the superior adaptability of the proposed method, achieving consistently high segmentation accuracy across domains. With only 5% annotation budget, our framework consistently outperforms state-of-the-art segmentation and domain adaptation methods, achieving robust segmentation accuracy across diverse domains and approaching the performance of fully supervised learning. Conclusions: The proposed method achieves superior accuracy in brain tumor region segmentation and precise boundary delineation under a limited annotation budget. It effectively mitigates domain shift while fully complying with data privacy regulations. Consequently, our framework relieves manual annotation bottlenecks and accelerates the cross-center deployment of accurate diagnostic tools, facilitating the clinical application of domain adaptation. Full article

Deep learning-based sheep face recognition technology significantly enhances the automation of individual sheep identification, providing critical technical support for smart livestock farming and precision agriculture. However, in real farming environments, factors such as complex backgrounds, illumination variations, and the high visual similarity of sheep faces severely constrain the comprehensive performance of recognition systems regarding accuracy and real-time capability. To address these challenges, we propose a cascaded framework comprising the WRT-DETR model for detection and LG-MobileViT for identification. WRT-DETR integrates multi-scale wavelet residual modeling and adaptive feature interaction into the RT-DETR architecture to effectively handle complex backgrounds. Subsequently, LG-MobileViT utilizes local–global collaborative modeling to distinguish fine-grained features while maintaining a lightweight footprint suitable for edge devices. Experiments conducted on a dataset of 400 individuals and 20,000 images demonstrate that WRT-DETR achieves 92.5% mAP50 in detection tasks. Furthermore, LG-MobileViT attains 98.97% recognition accuracy with a parameter size of only 4.57 MB. On edge computing platforms, the integrated system reaches an inference speed approaching 100 FPS. These results confirm that the proposed framework offers an efficient, reliable technical solution for non-contact, precise sheep identification in practical precision agriculture scenarios. Full article

Motor temperature prediction is critical for ensuring the reliability and safe operation of high-power-density electric drives. Since direct measurement of internal temperatures, especially rotor and magnet temperatures, is often impractical, virtual sensing has become an important research direction. This review provides a structured clarification of motor temperature prediction technologies. First, the physical foundations of motor thermal behavior are revisited, emphasizing multi-source loss generation, electro-thermal coupling mechanisms, and the dominant influence of time-varying boundary conditions. Second, existing estimation methodologies are systematically categorized into physics-based thermal models, observer- and identification-based approaches, and data-driven machine learning frameworks. Their mathematical principles, information bottlenecks, computational trade-offs, and deployment constraints are comparatively analyzed. Particular attention is given to hybrid and physics-informed methods, where reduced-order thermal networks, parameter adaptation, and learning-based residual correction are integrated to enhance robustness. Future developments should focus on uncertainty-aware estimation, lifecycle-adaptive modeling, and reliable temperature field inference under sparse sensing conditions. Full article