Search Results (1,837)

This case-based review examines the spectrum of leukemic ocular involvement, focusing on its prognostic implications. A rare case of relapsed acute myeloid leukemia (AML) in a 63-year-old man is presented, featuring simultaneous orbital proptosis, adnexal involvement, choroidal and retinal infiltration, and hemorrhagic changes affecting both the anterior and posterior segments. This constellation of findings, affecting multiple ocular structures concurrently, highlights the eye’s potential role as a sanctuary site for leukemic cells and underscores the diagnostic challenge of distinguishing direct infiltration from treatment-related or secondary vascular damage. This case, integrated with a literature review, emphasizes that multifocal ocular signs may serve as early indicators of leukemic relapse and reinforce the need for close collaboration between ophthalmologists and hematologists in guiding patient management. Full article

The objective of person re-identification (Re-ID) is to recognize a given target pedestrian across different cameras. However, perspective variations, resulting from differences in shooting angles, often significantly impact the accuracy of person Re-ID. To address this issue, this paper presents an attention-enhanced person Re-ID algorithm based on appearance–gait information interaction. Specifically, appearance features and gait features are first extracted from RGB images and gait energy images (GEIs), respectively, using two ResNet-50 networks. Then, a multimodal information exchange module based on the attention mechanism is designed to build a bridge for information exchange between the two modalities during the feature extraction process. This module aims to enhance the feature extraction ability through mutual guidance and reinforcement between the two modalities, thereby improving the model’s effectiveness in integrating the two types of modal information. Subsequently, to further balance the signal-to-noise ratio, importance weight estimation is employed to map perspective information into the importance weights of the two features. Finally, based on the autoencoder structure, the two features are weighted and fused under the guidance of importance weights to generate fused features that are robust to perspective changes. The experimental results on the CASIA-B dataset indicate that, under conditions of viewpoint variation, the method proposed in this paper achieved an average accuracy of 94.9%, which is 1.1% higher than the next best method, and obtained the smallest variance of 4.199, suggesting that the method proposed in this paper is not only more accurate but also more stable. Full article

This paper examines how marketization influences the spatial effects of household consumption upgrading in China, by analyzing provincial panel data from China between 2010 and 2022. The study employs a two-way fixed-effects Spatial Durbin Model to capture both the direct effects of marketization within a region and the spillover effects transmitted to neighboring regions. This model incorporates spatial dependence in both dependent and independent variables, providing a comprehensive assessment of spatial interactions. The results reveal that marketization and consumption upgrading both have the spatial pattern characteristics of significant spatial difference and agglomeration features. Marketization considerably encourages the upgrading of local people’s consumption and has positive spillover effects on the consumption upgrading levels of nearby regions. Mechanism analysis shows that market competition and enterprise innovation play key roles in this process. Heterogeneity analysis shows in eastern regions, areas with high industrial upgrading levels, high financial agglomeration levels, and high house prices, and the promotion effect of marketization on household consumption upgrading is more pronounced. These findings suggest that promoting differentiated regional marketization reforms, amplifying the spillover effects of marketization, reinforcing the dual engine of competition and innovation, and strengthening industrial upgrading and financial agglomeration are key to promote Chinese household consumption upgrading. Full article

Hierarchical Reinforcement Learning (HRL) is effective for long-horizon and sparse-reward tasks by decomposing complex decision processes, but its real-world application remains limited due to instability between levels, inefficient subgoal scheduling, delayed responses, and poor interpretability. To address these challenges, we propose Timed and Bionic Circuit Hierarchical Reinforcement Learning (TBC-HRL), a biologically inspired framework that integrates two mechanisms. First, a timed subgoal scheduling strategy assigns a fixed execution duration $\tau$ to each subgoal, mimicking rhythmic action patterns in animal behavior to improve inter-level coordination and maintain goal consistency. Second, a Neuro-Dynamic Bionic Circuit Network (NDBCNet), inspired by the neural circuitry of C. elegans, replaces conventional fully connected networks in the low-level controller. Featuring sparse connectivity, continuous-time dynamics, and adaptive responses, NDBCNet models temporal dependencies more effectively while offering improved interpretability and reduced computational overhead, making it suitable for resource-constrained platforms. Experiments across six dynamic and complex simulated tasks show that TBC-HRL consistently improves policy stability, action precision, and adaptability compared with traditional HRL, demonstrating the practical value and future potential of biologically inspired structures in intelligent control systems. Full article

We present a bibliometric analysis of unmanned aerial vehicle (UAV) research that replaces simple keyword filtering with a context-aware, two-tier pipeline. Records from Web of Science and Scopus (198,152 total) were harmonized and de-duplicated in three stages (DOI, normalized title, blockwise Jaro–Winkler), yielding 129,124 unique items. To separate UAV work from entomology using overlapping vocabulary (e.g., swarm), we first applied rule-based weak labels with explicit UAV and insect regex families and a UAV context rule for “swarm,” then trained an elastic-net logistic regression on TF–IDF features and tuned the decision threshold to meet a high-precision target on a held-out split. The final corpus comprises 129,099 UAV records. Beyond lexical inventories, a keyword co-occurrence timeline shows reinforcement learning increasingly aligned with path planning and collision avoidance, while constraints such as energy and communication persist. A co-authorship network reveals bridging authors that connect guidance/control, perception, and communication subfields. The results show how UAV research is organized around central scientific problems and identify persistent obstacles such as energy efficiency, communication reliability, and robust decision-making in dynamic conditions. Full article

To address circularity and resource recovery in modern structural applications, industry is seeking materials that are sustainable and lightweight. Although natural fiber-reinforced composites offer sustainability advantages, their mechanical properties remain inferior to those of synthetic fiber systems, limiting practical deployment. Flax fibers were selected as reinforcement due to their high specific stiffness, biodegradability, and wide availability. This study implements a three-level strategy to enhance the flexural performance of flax fiber-reinforced composites: at the process level, curing under distinct heating rates to promote a more uniform polymer network; at the material level, incorporation of a carbonaceous additive derived from fuel–oil furnace waste to strengthen interfacial adhesion; and at the structural level, adoption of a sandwich configuration with a recycled PET core to increase section bending inertia. Specimens were fabricated via vacuum-assisted resin transfer molding (VARTM) and tested using a three-point bending method. Mechanical testing shows clear improvements in flexural performance, with the sandwich architecture yielding the highest values and increasing flexural strength by up to 4.52× relative to the other conditions. For the curing series, FTIR indicates greater reaction extent, evidenced by lower intensities of the epoxide ring at 915 cm⁻¹ and glycidyl/oxirane band near 972 cm⁻¹, together with a more pronounced C–O–C stretching region, consistent with the higher flexural response. While SEM observations revealed interfacial debonding at 5% FCB, a hybrid mechanism with crack deflection appeared at 10%. This transition created tortuous crack paths, consistent with the higher flexural strength and modulus at 10% FCB. A distinctive feature of this work is the integration of three reinforcement strategies—controlled curing, waste-derived carbon additive, and recycled PET sandwich design. This integration not only enhances the performance of natural fiber composites but also emphasizes sustainability by valorizing recycled and waste-derived resources, thereby supporting the development of greener composite materials. Full article

Deep neural network-based approaches have obtained remarkable progress in monaural speech enhancement. Nevertheless, current cutting-edge approaches remain vulnerable to complex acoustic scenarios. We propose a Symmetric Combined Convolution Network with ConvLSTM (SCCN) for monaural speech enhancement. Specifically, the Combined Convolution Block utilizes parallel convolution branches, including standard convolution and two different depthwise separable convolutions, to reinforce feature extraction in depthwise and channelwise. Similarly, Combined Deconvolution Blocks are stacked to construct the convolutional decoder. Moreover, we introduce the exponentially increasing dilation between convolutional kernel elements in the encoder and decoder, which expands receptive fields. Meanwhile, the grouped ConvLSTM layers are exploited to extract the interdependency of spatial and temporal information. The experimental results demonstrate that the proposed SCCN method obtains on average 86.00% in STOI and 2.43 in PESQ, which outperforms the state-of-the-art baseline methods, confirming the effectiveness in enhancing speech quality. Full article

Carbon Fiber-Reinforced Polymer (CFRP) composites, widely used across industries, exhibit various damage mechanisms depending on the loading conditions applied. This study employs a structural health monitoring (SHM) approach to investigate the three primary failure modes, fiber breakage, matrix cracking, and delamination, in thermoset quasi-isotropic CFRPs subjected to quasi-static tensile loading until failure. Acoustic emission (AE) signals acquired from an experiment were leveraged to analyze and classify these real-time signals into the failure modes using machine learning (ML) techniques. Due to the extensive number of AE signals recorded during testing, manually classifying these failure mechanisms through waveform inspection was impractical. ML, alongside ensemble learning, algorithms were implemented to streamline the classification, making it more efficient, accurate, and reliable. Conventional AE parameters from the data acquisition system and feature extraction techniques applied to the recorded waveforms were implemented exclusively as classification features to investigate their reliability and accuracy in classifying failure modes in CFRPs. The classification models exhibited up to 99% accuracy, as depicted by evaluation metrics. Further studies, using cross-correlation techniques, ascertained the presence of fiber break events occurring in the bundles as the thermoset CFRP composite approached failure. These findings highlight the significance of integrating machine learning into SHM for the early detection of real-time damage and effective monitoring of residual life in composite materials. Full article

The segmentation and classification of color are crucial stages in image processing, computer vision, and pattern recognition, as they significantly impact the results. The diverse, hand-labeled datasets in the literature are applied for monochromatic or color segmentation in specific domains. On the other hand, synthetic datasets are generated using statistics, artificial intelligence algorithms, or generative artificial intelligence (AI). This last one includes Large Language Models (LLMs), Generative Adversarial Neural Networks (GANs), and Variational Autoencoders (VAEs), among others. In this work, we propose VitralColor-12, a synthetic dataset for color classification and segmentation, comprising twelve colors: black, blue, brown, cyan, gray, green, orange, pink, purple, red, white, and yellow. VitralColor-12 addresses the limitations of color segmentation and classification datasets by leveraging the capabilities of LLMs, including adaptability, variability, copyright-free content, and lower-cost data—properties that are desirable in image datasets. VitralColor-12 includes pixel-level classification and segmentation maps. This makes the dataset broadly applicable and highly variable for a range of computer vision applications. VitralColor-12 utilizes GPT-5 and DALL·E 3 for generating stained-glass images. These images simplify the annotation process, since stained-glass images have isolated colors with distinct boundaries within the steel structure, which provide easy regions to label with a single color per region. Once we obtain the images, we use at least one hand-labeled centroid per color to automatically cluster all pixels based on Euclidean distance and morphological operations, including erosion and dilation. This process enables us to automatically label a classification dataset and generate segmentation maps. Our dataset comprises 910 images, organized into 70 generated images and 12 pixel segmentation maps—one for each color—which include 9,509,524 labeled pixels, 1,794,758 of which are unique. These annotated pixels are represented by RGB, HSL, CIELAB, and YCbCr values, enabling a detailed color analysis. Moreover, VitralColor-12 offers features that address gaps in public resources such as violin diagrams with the frequency of colors across images, histograms of channels per color, 3D color maps, descriptive statistics, and standardized metrics, such as ΔE76, ΔE94, and CIELAB Chromacity, which prove the distribution, applicability, and realistic perceptual structures, including warm, neutral, and cold colors, as well as the high contrast between black and white colors, offering meaningful perceptual clusters, reinforcing its utility for color segmentation and classification. Full article

Wire Arc Additive Manufacturing (WAAM) has been recognized as an efficient and cost-effective metal additive manufacturing technique due to its high deposition rate and scalability for large components. However, the quality and repeatability of WAAM parts are highly sensitive to process parameters such as arc voltage, current, wire feed rate, and torch travel speed, requiring advanced monitoring and adaptive control strategies. In this study, a vision-based monitoring system integrated with a reinforcement learning framework was developed to enable intelligent in situ control of WAAM. A custom optical assembly employing mirrors and a bandpass filter allowed simultaneous top and side views of the melt pool, enabling real-time measurement of layer height and width. These geometric features provide feedback to a tabular Q-learning algorithm, which adaptively adjusts voltage and wire feed rate through direct hardware-level control of stepper motors. Experimental validation across multiple builds with varying initial conditions demonstrated that the RL controller stabilized layer geometry, autonomously recovered from process disturbances, and maintained bounded oscillations around target values. While systematic offsets between digital measurements and physical dimensions highlight calibration challenges inherent to vision-based systems, the controller consistently prevented uncontrolled drift and corrected large deviations in deposition quality. The computational efficiency of tabular Q-learning enabled real-time operation on standard hardware without specialized equipment, demonstrating an accessible approach to intelligent process control. These results establish the feasibility of reinforcement learning as a robust, data-efficient control technique for WAAM, capable of real-time adaptation with minimal prior process knowledge. With improved calibration methods and expanded multi-physics sensing, this framework can advance toward precise geometric accuracy and support broader adoption of machine learning-based process monitoring and control in metal additive manufacturing. Full article

The rapid development of the Internet of Things (IoT) poses significant problems in securing heterogeneous, massive, and high-volume network traffic against cyber threats. Traditional intrusion detection systems (IDSs) are often found to be poorly scalable, or are ineffective computationally, because of the presence of redundant or irrelevant features, and they suffer from high false positive rates. Addressing these limitations, this study proposes a hybrid intelligent model that combines quantum computing, chaos theory, and deep learning to achieve efficient feature selection and effective intrusion classification. The proposed system offers four novel modules for feature optimization: chaotic swarm intelligence, quantum diffusion modeling, transformer-guided ranking, and multi-agent reinforcement learning, all of which work with a graph-based classifier enhanced with quantum attention mechanisms. This architecture allows as much as 75% feature reduction, while achieving 4% better classification accuracy and reducing computational overhead by 40% compared to the best-performing models. When evaluated on benchmark datasets (NSL-KDD, CICIDS2017, and UNSW-NB15), it shows superior performance in intrusion detection tasks, thereby marking it as a viable candidate for scalable and real-time IoT security analytics. Full article

To investigate the influence of laser cladding overlap rate on the microstructure and corrosion resistance of cladded layers, Ni60-WC composite coatings with different overlap rates (30%, 50%, and 70%) were prepared on E690 offshore steel in this study. The relationship between the corrosion resistance and microstructure of the cladded layers fabricated at different overlap rates was analyzed using an electrochemical workstation, scanning electron microscope, X-ray diffractometer, and energy dispersive spectrometer. The results demonstrate that the overlap rate exerts a significant impact on the corrosion resistance of the cladded layers, and the corrosion resistance of the cladded layers gradually improves with the increase in overlap rate. The cladded layer prepared with a 70% overlap rate exhibits excellent corrosion resistance, featuring the highest open-circuit potential (−0.31 V vs. SCE), the lowest corrosion current density (3.35 μA/cm²), the largest capacitive arc radius in the electrochemical impedance spectroscopy (EIS), and a relatively flat surface after corrosion tests. Microstructural characterization results indicate that the increase in overlap rate promotes grain refinement and the formation of reinforcing phases (e.g., M₂₃C₆). The coating with a 70% overlap rate possesses the densest microstructure and abundant flocculent carbides, which act as an effective barrier against the penetration of corrosive media, thereby endowing it with optimal performance. Full article

This paper proposes an interpretable optimization method for power grid outage scheduling based on reinforcement learning. An outage scheduling optimization model is proposed, considering the convergence of power flow calculation, voltage violations, and operational economic behavior as objectives, while considering constraints such as simultaneous outage constraints, mutually exclusive constraints, and maintenance windows. Key features of the outage schedule are selected based on Shapley values to construct a Markov optimization model for outage scheduling. A deep reinforcement learning agent is established to optimize the outage schedule. The proposed method is applied to the IEEE-39 and IEEE-118 bus system for validation. Experimental results show that the proposed method outperforms existing algorithms in terms of voltage violation, total power losses, and computational time. The proposed method eliminates all voltage violations and reduces active power losses up to 5.7% and computation time by 6.8 h compared to conventional heuristic algorithms. Full article

The comprehensive renovation of existing buildings has become imperative and is recognized as a central priority within the European Union’s agenda (European Green Deal). The objectives of this initiative include reducing energy consumption, mitigating environmental pollution, and achieving long-term decarbonization targets. This research addresses the case of load-bearing masonry buildings constructed in the post-World War II period, characterized by specific geometric and volumetric features. Current regulations on seismic design and thermal protection reveal significant deficiencies in both the structural safety and the energy performance of these buildings. Recent seismic events and the increasing demand for electricity further highlight the urgency of integrated retrofitting measures that simultaneously enhance structural resistance and improve thermal protection. This research aims to develop an integrated retrofitting approach that simultaneously improves seismic resistance and energy efficiency. A review of strengthening techniques and thermal upgrades was carried out, followed by a critical assessment of their applicability. The proposed intervention combines two comparable seismic reinforcement schemes with thermal improvements, implemented through a one-sided reinforced cement mortar overlay coupled with external thermal insulation materials. Analyses demonstrate that the retrofit increases the structural resistance to a_gR = 0.10 g and upgrades the building envelope to current energy efficiency requirements. The results confirm that the method is both effective and feasible, offering a replicable solution for similar residential masonry buildings. This study concludes that integrated retrofitting can extend building service life, enhance occupant safety and comfort, and provide a practical framework for large-scale application in sustainable renovation practices, which is especially significant for Serbia and other Balkan countries, considering that the analyzed case study buildings are characteristic representatives for these regions. Full article

Offline reinforcement learning suffers from critical vulnerability to data corruption from sensor noise or adversarial attacks. Recent research has achieved a lot by downweighting corrupted samples and fixing the corrupted data, while data corruption induces feature entanglement that undermines policy robustness. Existing methods fail to identify causal features behind performance degradation caused by corruption. To analyze causal relationships in corrupted data, we propose a method, Robust Causal Feature Disentanglement(RCFD). Our method introduces a learnable causal feature disentanglement mechanism specifically designed for reinforcement learning scenarios, integrating the CausalVAE framework to disentangle causal features governing environmental dynamics from corruption-sensitive non-causal features. Theoretically, this disentanglement confers a robustness advantage under data corruption conditions. Concurrently, causality-preserving perturbation training injects Gaussian noise solely into non-causal features to generate counterfactual samples and is enhanced by dual-path feature alignment and contrastive learning for representation invariance. A dynamic graph diagnostic module further employs graph convolutional attention networks to model spatiotemporal relationships and identify corrupted edges through structural consistency analysis, enabling precise data repair. The results exhibit highly robust performance across D4rl benchmarks under diverse data corruption conditions. This confirms that causal feature invariance helps bridge distributional gaps, promoting reliable deployment in complex real-world settings. Full article