Search Results (1,425)

Automatic Target Recognition (ATR) in single-view sonar imagery is severely hampered by geometric distortions, acoustic shadows, and incomplete target information due to occlusions and the slant-range imaging geometry, which frequently give rise to misclassification and hinder practical underwater detection applications. To address these critical limitations, this paper proposes a Multi-View Deep Convolutional Neural Network (MVDCNN) based on feature-level fusion for robust sonar image target recognition. The MVDCNN adopts a highly modular and extensible architecture consisting of four interconnected modules: an input reshaping module that adapts multi-view images to match the input format of pre-trained backbone networks via dimension merging and channel replication; a shared-weight feature extraction module that leverages Convolutional Neural Network (CNN) or Transformer backbones (e.g., ResNet, Swin Transformer, Vision Transformer) to extract discriminative features from each view, ensuring parameter efficiency and cross-view feature consistency; a feature fusion module that aggregates complementary features (e.g., target texture and shape) across views using max-pooling to retain the most salient characteristics and suppress noisy or occluded view interference; and a lightweight classification module that maps the fused feature representations to target categories. Additionally, to mitigate the data scarcity bottleneck in sonar ATR, we design a multi-view sample augmentation method based on sonar imaging geometric principles: this method systematically combines single-view samples of the same target via the combination formula and screens valid samples within a predefined azimuth range, constructing high-quality multi-view training datasets without relying on complex generative models or massive initial labeled data. Comprehensive evaluations on the Custom Side-Scan Sonar Image Dataset (CSSID) and Nankai Sonar Image Dataset (NKSID) demonstrate the superiority of our framework over single-view baselines. Specifically, the two-view MVDCNN achieves average classification accuracies of 94.72% (CSSID) and 97.24% (NKSID), with relative improvements of 7.93% and 5.05%, respectively; the three-view MVDCNN further boosts the average accuracies to 96.60% and 98.28%. Moreover, MVDCNN substantially elevates the precision and recall of small-sample categories (e.g., Fishing net and Small propeller in NKSID), effectively alleviating the class imbalance challenge. Mechanism validation via t-Distributed Stochastic Neighbor Embedding (t-SNE) feature visualization and prediction confidence distribution analysis confirms that MVDCNN yields more separable feature representations and more confident category predictions, with stronger intra-class compactness and inter-class discrimination in the feature space. The proposed MVDCNN framework provides a robust and interpretable solution for advancing sonar ATR and offers a technical paradigm for multi-view acoustic image understanding in complex underwater environments. Full article

Volume electron microscopy (Volume-EM) has transformed structural cell biology by enabling nanometre-resolution imaging across cellular and tissue scales. Serial-section TEM, Serial Block-Face Scanning Electron Microscopy (SBF-SEM), Focused Ion Beam Scanning Electron Microscopy (FIB-SEM) and multi-beam SEM now routinely generate terabyte-scale volumes that capture organelles, synapses and neural circuits in three dimensions, while cryogenic Volume-EM extends this landscape by preserving vitrified, fully hydrated specimens in a near-native state. Together, these room-temperature and cryogenic modalities define a continuum of approaches that trade off volume, resolution, throughput and structural fidelity, and increasingly interface with correlative light microscopy and cryo-electron tomography. In parallel, advances in computation have turned Volume-EM into a data-intensive discipline. Multistage preprocessing pipelines for alignment, denoising, stitching and intensity normalisation feed into automated segmentation frameworks that combine convolutional neural networks, affinity-based supervoxel agglomeration, flood-filling networks and, more recently, diffusion-based generative restoration. Weakly supervised and self-supervised learning, multi-task objectives and human-AI co-training mitigate the scarcity of dense ground truth, while distributed storage and streaming inference architectures support segmentation and proofreading at the terascale and beyond. Open resources such as COSEM, MICRONS, OpenOrganelle and EMPIAR provide benchmark datasets, interoperable file formats and reference workflows that anchor method development and cross-laboratory comparison. In this review, we first outline the physical principles and imaging modes of conventional and cryogenic Volume-EM, then describe current best practices in data acquisition and preprocessing, and finally survey the emerging ecosystem of AI-driven segmentation and analysis. We highlight how cryo-Volume-EM expands the field towards native-state structural biology, and how multimodal integration with light microscopy, cryo-electron tomography (cryo-ET) and spatial omics is pushing Volume-EM from descriptive imaging towards predictive, mechanistic, cross-scale models of cell physiology, disease ultrastructure and neural circuit function. Full article

Glass Fiber Reinforced Polymer (GFRP)-reinforced concrete beams have gained significant prominence in structural engineering due to their advantageous mechanical and durability characteristics. However, the influence of training–testing data partitioning on machine learning (ML)-based deflection prediction for such members remains insufficiently explored. This study addresses this gap by evaluating the predictive performance of the K-Nearest Neighbors (KNN) regression algorithm in estimating the load–deflection behavior of GFRP-reinforced high-strength concrete beams. The experimental program comprised nine beams manufactured with concrete strength classes C45, C50, and C65, followed by ML-based deflection analyses using multiple data-splitting strategies. Findings indicate that the KNN model employing an 80:20 training–testing ratio provides the most accurate deflection predictions, achieving approximately 80% agreement with experimental results, while a higher prediction accuracy of approximately 85% was observed for beams with the highest concrete compressive strength (C65). Experimentally recorded deflections ranged from approximately 20 mm to values exceeding 50 mm, depending on the concrete strength class and loading level. Owing to its superior performance, the KNN model with an 80:20 training–testing ratio is recommended for predicting the deflection capacities of GFRP-reinforced high-strength concrete members. The study further examined the structural response associated with the use of GFRP as longitudinal tensile reinforcement. A consistent failure mechanism was observed across all beams, characterized by the formation of a single, wide vertical crack initiating at the beam’s soffit, regardless of concrete strength class. These observations contribute to a deeper understanding of the flexural behavior and fracture characteristics of GFRP-reinforced high-strength concrete beams and provide a foundation for future modeling efforts. Full article

Aerial situational awareness (SA) faces significant challenges due to inherent complexity involving large-scale dynamic entities and intricate spatio-temporal relationships. While deep learning advances SA for specific data modalities (static or time-series), existing approaches often lack the holistic, vision-centric perspective essential for human decision-making. To bridge this gap, we propose a unified GNN-CV framework for operational-level SA. This framework leverages mature computer vision (CV) architectures to intelligently process radar-map-like representations, addressing diverse SA tasks within a unified paradigm. Key innovations include methods for sparse entity attribute transformation graph neural networks (SET-GNNs), large-scale radar map reconstruction, integrated feature extraction, specialized two-stage pre-training, and adaptable downstream task networks. We rigorously evaluate the framework on critical operational-level tasks: aerial swarm partitioning and configuration recognition. The framework achieves an impressive end-to-end recognition accuracy exceeding 90.1%. Notably, in specialized tactical scenarios featuring small, large, and irregular flight intervals within formations, configuration recognition accuracy surpasses 85.0%. Even in the presence of significant position and heading disturbances, accuracy remains above 80.4%, with millisecond response cycles. Experimental results highlight the benefits of leveraging mature CV techniques such as image classification, object detection, and image generation, which enhance the efficacy, resilience, and coherence of intelligent situational awareness. Full article

In this study, regression-based machine learning models were developed to predict the melt pool width and depth formed during the Laser Powder Bed Fusion (LPBF) process for Fe-3.4Si and Fe-6Si alloys. Based on experimentally obtained melt pool width and depth data, a total of 11 regression models were trained and evaluated, and hyperparameters were optimized via Bayesian optimization. Key process parameters were identified through data preprocessing and feature engineering, and SHAP analysis confirmed that the input energy had the strongest influence on both melt pool width and depth. The comparison of prediction performance revealed that the support vector regressor with a linear kernel (SVR_lin) exhibited the best performance for predicting melt pool width, while the multilayer perceptron (MLP) model achieved the best results for predicting melt pool depth. Based on these trained models, a power–velocity (P-V) process map was constructed, incorporating boundary conditions such as the overlap ratio and the melt pool morphology. The optimal input energy range was derived as 0.45 to 0.60 J/mm, ensuring stable melt pool formation. Specimens manufactured under the derived conditions were analyzed using 3D X-ray CT, revealing porosity levels ranging from 0.29% to 2.89%. In particular, the lowest porosity was observed under conduction mode conditions when the melt pool depth was approximately 1.0 to 1.5 times the layer thickness. Conversely, porosity tended to increase in the transition mode and lack of fusion regions, consistent with the model predictions. Therefore, this study demonstrated that a machine learning-based regression model can reliably predict melt pool characteristics in the LPBF process of Fe-Si alloys, contributing to the development of process maps and optimization strategies. Full article

Background/Objectives: Late-life depression is prevalent yet frequently underdiagnosed, underscoring the need for accessible and safe non-pharmacological approaches. Pet-assisted interventions, including live animal-assisted therapy and robotic pets, have gained attention, but their comparative effectiveness remains unclear. This study aimed to evaluate and rank different pet-assisted approaches for reducing depressive symptoms in older adults using network meta-analysis. Methods: We systematically searched PubMed, Embase, Web of Science, and the Cochrane Library up to August 2025 for randomized controlled trials involving adults aged 60 years or older with depression. The protocol was prospectively registered on INPLASY (INPLASY2025100023). Depression severity, assessed using validated scales, was synthesized using a frequentist random-effects network meta-analysis framework. Results: Twenty trials involving 1073 participants were included. Live animal-assisted therapy produced the greatest reduction in depressive symptoms versus passive control (SMD −2.04; 95% CI −3.03 to −1.04). Combining it with gait training (structured walking-based activity conducted with the animal) was associated with a reduction in depressive symptoms (SMD −4.82; 95% CI −6.69 to −2.95). Robotic pets showed a directionally beneficial but non-significant effect (SMD −1.21; 95% CI −2.79 to 0.38). Conclusions: Pet-assisted interventions are effective in reducing depressive symptoms among older adults. Live animal-assisted therapy, particularly when delivered in structured or combined formats, shows the greater benefit. Robotic pets may serve as a practical alternative when live animals cannot be implemented. Full article

The significant transformation of agricultural production and operation models has reshaped the supply-demand structure of rural land, providing growth opportunities for new agricultural management entities characterized by large-scale operation. Their large-scale land demand has directly driven an upward trend in the transfer prices of contracted land management rights. By analyzing this practical phenomenon, this study explores the intrinsic logic behind the rising transfer prices of contracted land management rights under the participation of new agricultural management entities, aiming to provide references for further regulating the formation mechanism of transfer prices and promoting the healthy development of the land transfer market. Based on the sample survey data of farmers from the Songnen Plain and Sanjiang Plain in Northeast China, this study adopts the cluster-robust Ordinary Least Squares (OLS) model and moderating effect model for analysis. The results show that the participation of new agricultural management entities exerts a positive impact on the transfer price of contracted land management rights; the impact of new agricultural management entities’ participation on the transfer price is positively moderated by agricultural production efficiency; and the impact also presents heterogeneity across different villages and land parcels. Compared with remote villages and paddy parcels, the participation of new agricultural management entities has a more significant impact on the transfer price of contracted land management rights in township-adjacent villages and dryland parcels. Therefore, to reasonably standardize the transfer price of contracted land management rights, efforts should focus on further strengthening policy guidance to standardize the participation mechanism of new agricultural management entities, regulating the transfer market to establish a dynamic monitoring mechanism for transfer prices, and strengthening the training and guidance for new agricultural management entities to connect and drive farmers so as to improve the agricultural production efficiency of individual farmers. Full article

Background/Objectives: Inborn errors of metabolism (IEM) are chronic, life-threatening genetic disorders with a significant cumulative prevalence worldwide. Advances in early diagnosis and treatment have significantly increased life expectancy, underscoring the need for specialised adult care units and the establishment of structured transition programmes from paediatric to adult services. We hereby present a functional transition model for IEM patients and share our implementation experience. Methods: Initiated in 2012, the partnership between the paediatric Hospital Sant Joan de Déu (HSJD) and the adult-care centre at Hospital Clinic of Barcelona (HCB) culminated in 2019 with the transference of the first IEM patients under the structured A10! Programme. This model is structured around the transition units of paediatric and adult centres to guarantee communication and functional management. Regular monthly meetings at each centre and joint quarterly sessions allowed for protocol harmonisation and personalised care planning. Coordinated engagement of the multidisciplinary health care teams with patients and families smoothed the transfer process. Results: Between 2019 and 2024, 94 IEM patients were successfully transferred. Diagnoses included intermediary metabolism defects (71.23%), lipid metabolism and transport disorders (4.25%), heterocyclic compound metabolism (2.12%), complex molecules and organelle dysfunction (6.37%), cofactor and mineral metabolism (2.12%), signalling defects (5.31%), and unclassified cases (8.51% of rare disorders, maybe non-IEM). Transition formats included 21 in-person joint visits in HSJD, 37 remote transitions during the COVID-19 pandemic, and 36 streamlined transfers via standardised protocols. Sessions, trainings, and meetings allowed the exchange of patients’ needs and protocols. Conclusions: The successful transference of IEM patients requires structured programmes with interdisciplinary paediatric and adult teams, joining efforts with the patient, families, and caregivers. Communication between paediatric and adult transition units is essential to promote continuity of care and patient empowerment. While constantly updated, this model has proven effective, gaining positive evaluations from healthcare professionals and patients alike, representing a scalable framework for lifelong management of IEM in adult care settings. Full article

Dispersed data classification presents significant challenges due to structural variations, restricted information exchange, and the need for powerful decision-making strategies. This study introduces a dynamic classification system based on coalition formation using local models trained on independently collected local data. We explore two distinct coalition strategies: unified coalitions, which group models with similar prediction behaviors, and diverse coalitions, which aggregate models exhibiting contrasting decision tendencies. The impact of weighted and unweighted prediction aggregation is also examined to determine the influence of model reliability on global decision-making. Our framework uses Pawlak’s conflict analysis to form optimal coalitions. Experimental evaluations using multiple datasets demonstrate that coalition-based approaches significantly improve classification accuracy compared to operating individual models. The weighted diverse coalitions produce the most stable results. Statistical analyses confirm the effectiveness of the proposed methodology, highlighting the advantages of adaptive coalition strategies in dispersed environments. Full article

Background: Digitalisation is transforming medical education, but its integration into ultrasound training remains limited. This study evaluates the needs of students and physicians regarding digitally supported ultrasound education. Materials and Methods: A multi-year cross-sectional study (2017–2022) employed two standardised questionnaires. The first assessed the perceived relevance of ultrasound in medical education, the desirability of compulsory teaching, and the integration of digital media and case-based learning. The second explored user-centred requirements for e-learning formats, including functionality, multimedia design, usability, interactivity, and financing, as well as current use of digital devices and reference materials. Data were collected using dichotomous and 7-point Likert scales (1 = high need/strong agreement, 7 = low need/weak agreement). Results: A total of 3479 responses were analysed (2821 students; 658 physicians). Both groups showed strong support for integrating ultrasound into curricula (1.3 ± 0.7) and mandatory education (1.4 ± 0.9), with students expressing significantly greater support (p < 0.001). There was broad agreement on the integration and development of digital media (1.7 ± 1.0), as well as the use of case studies (1.4 ± 0.8), with no significant differences between groups (p > 0.05). Case-based learning as a stand-alone format was less favoured (3.4 ± 1.9). In the user-centred needs analysis, both groups rated features like search functions (1.4 ± 0.8), usability (1.5 ± 0.9), and learning objective checks (2.7 ± 1.6) as important. High-quality media (1.5 ± 0.9) and pathology explanations (1.6 ± 1.1) were also highly valued. Students primarily relied on digital platforms, while physicians used a more varied mix of digital platforms, guidelines, and textbooks. Conclusions: The study highlights the need for more extensive, digitally supported ultrasound training, with a focus on functionality and usability. Standardisation through structured certification processes should be considered for future implementation. Full article

Accurate group target association is essential for multi-sensor multi-target tracking, particularly in heterogeneous radar systems where systematic biases, asynchronous observations, and dense formations frequently cause ambiguous or incorrect associations. Existing approaches often rely on strict spatial assumptions or pre-trained models, limiting their robustness when measurement distortions and sensor-specific deviations are present. To address these challenges, this work proposes a robust association framework that integrates deep feature embedding, density-adaptive clustering, and global graph-theoretic matching. The method first applies an autoencoder–HDBSCAN clustering scheme to extract stable latent representations and obtain adaptive group structures under nonlinear distortions and non-uniform target densities. A weighted bipartite graph is then constructed, and a global optimal matching strategy is employed to compensate for heterogeneous systematic errors while preserving inter-group structural consistency. A mutual-support verification mechanism further enhances robustness against random disturbances. Monte Carlo experiments show that the proposed method maintains over 90% association accuracy even in dense scenarios with a target spacing of 1.4 km. Under various systematic bias conditions, it outperforms representative baselines such as Deep Association and JPDA by more than 20%. These results demonstrate the method’s robustness, adaptability, and suitability for practical multi-radar applications. The framework is training-free and easily deployable, offering a reliable solution for group target association in real-world multi-sensor fusion systems. Full article

The dual imperative of mitigating carbon emissions and maximizing hydrocarbon recovery has amplified global interest in carbon capture, utilization, and storage (CCUS) technologies. These integrated processes hold significant promise for achieving net-zero targets while extending the productive life of mature oil reservoirs. However, their effectiveness hinges on a nuanced understanding of the complex interactions between geological formations, reservoir characteristics, and injection strategies. In this study, a comprehensive machine learning-based framework is presented for estimating CO₂ storage capacity and enhanced oil recovery (EOR) performance simultaneously in subsurface reservoirs. The methodology combines simulation-driven uncertainty quantification with supervised machine learning to develop predictive surrogate models. Simulation results were used to generate a diverse dataset of reservoir and operational parameters, which served as inputs for training and testing three machine learning models: Random Forest, Extreme Gradient Boosting (XGBoost), and Artificial Neural Networks (ANN). The models were trained to predict three key performance indicators (KPIs): cumulative oil production (bbl), oil recovery factor (%), and CO₂ sequestration volume (SCF). All three models exhibited exceptional predictive accuracy, achieving coefficients of determination (R²) greater than 0.999 across both training and testing datasets for all KPIs. Specifically, the Random Forest and XGBoost models consistently outperformed the ANN model in terms of generalization, particularly for CO₂ sequestration volume predictions. These results underscore the robustness and reliability of machine learning models for evaluating and forecasting the performance of CO₂-EOR and sequestration strategies. To enhance model interpretability and support decision-making, SHapley Additive exPlanations (SHAP) analysis was applied. SHAP, grounded in cooperative game theory, offers a model-agnostic approach to feature attribution by assigning an importance value to each input parameter for a given prediction. The SHAP results provided transparent and quantifiable insights into how geological and operational features such as porosity, injection rate, water production rate, pressure, etc., affect key output metrics. Overall, this study demonstrates that integrating machine learning with domain-specific simulation data offers a scalable approach for optimizing CCUS operations. The insights derived from the predictive models and SHAP analysis can inform strategic planning, reduce operational uncertainty, and support more sustainable oilfield development practices. Full article

Aquaculture monitoring increasingly relies on computer vision to evaluate fish behavior and welfare under farming conditions. This dataset was collected in a commercial recirculating aquaculture system (RAS) integrated with hydroponics in Queretaro, Mexico, to support the development of robust visual models for Nile tilapia (Oreochromis niloticus). More than ten hours of underwater recordings were curated into 31 clips of 30 s each, a duration selected to balance representativeness of fish activity with a manageable size for annotation and training. Videos were captured using commercial action cameras at multiple resolutions (1920 × 1080 to 5312 × 4648 px), frame rates (24–60 fps), depths, and lighting configurations, reproducing real-world challenges such as turbidity, suspended solids, and variable illumination. For each recording, physicochemical parameters were measured, including temperature, pH, dissolved oxygen and turbidity, and are provided in a structured CSV file. In addition to the raw videos, the dataset includes 3520 extracted frames annotated using a polygon-based JSON format, enabling direct use for training object detection and behavior recognition models. This dual resource of unprocessed clips and annotated images enhances reproducibility, benchmarking, and comparative studies. By combining synchronized environmental data with annotated underwater imagery, the dataset contributes a non-invasive and versatile resource for advancing aquaculture monitoring through computer vision. Full article

Accurate and timely identification of cutter anomalies is crucial for ensuring the safety and efficiency of shield tunneling. To address the issues of poor timeliness and high costs associated with traditional periodic manual inspection methods, this study establishes a cutter anomaly identification model based on the BO-Light GBM algorithm, focusing on slightly weathered metamorphic rock formations. Six parameters closely related to the tunneling state were selected to construct the feature set, and one-class support vector machines (SVMs) were employed to remove anomalous samples. On this basis, a baseline Light GBM model with preset hyperparameters was developed, achieving a preliminary accuracy of 96.04%. Further hyperparameter tuning using Bayesian optimization boosted the overall accuracy of the final BO-Light GBM model to 99.40% while improving training efficiency by approximately 50% compared to exhaustive grid search. Interpretability analysis conducted via SHAP values revealed that chamber pressure, cutterhead rotation speed, total thrust, and cutterhead torque were the primary contributing features, with patterns consistent with actual tunneling conditions, confirming the accuracy of the model’s predictions. The research outcomes provide valuable theoretical guidance and technical support for similar engineering applications. Full article

Research in “cybertherapy” has explored innovative ways to integrate new technologies as innovative tools in psychological treatment, such as virtual reality. Avatars, as digital representations of users within virtual environments, represent an interesting tool for psychotherapists: they could be used to assess aspects of patients’ self-representations (assessment), to promote behavioral change based on an alternative self-image (treatment), or to exercise therapists’ skills in diagnosis and assessment (formation). Yet, the use of avatars in psychotherapy is still not widespread. In the present study, 77 certified psychotherapists evaluated the three possible uses of avatars described above in terms of technology acceptance model (TAM) factors: perceived usefulness, perceived ease of use and intention-to-use. Partially confirming the TAM, the results show that perceived usefulness in particular is an effective predictor of intention to use avatars in psychotherapy for all three possible uses. Attitudes towards avatars as a psychotherapeutic tool were slightly influenced by mental health professionals’ methodological approach, with cognitive-behavioral psychotherapists showing more positive attitudes towards avatars as a training tool. On the other hand, previous experiences with other technologies (e.g., conducting therapy online or not) affected the perception of avatars’ ease of use as a treatment tool. The present study contributes to identifying factors that influence mental health professionals’ attitudes towards technological innovations in the psychotherapy profession, giving directions for future research in cybertherapy adoption. Full article