Search Results (266)

High-precision identification and three-dimensional (3D) positioning of cabbage plants across their entire growth cycle are fundamental prerequisites for automated agricultural management. To overcome field challenges like extreme morphological variations, severe leaf occlusion, and bounding box jitter, we introduce a camera-LiDAR fusion perception system. First, an advanced SCEW-YOLOv8 architecture is proposed, sequentially integrating SPD-Conv downsampling, a C2f-CX global feature enhancement module, an EMA cross-space attention mechanism, and the WIoU v3 loss function. Evaluated on a comprehensive whole-growth-cycle cabbage dataset, the model achieves 95.8% mAP@0.5 and 90.8% recall with a real-time inference speed of 64.2 FPS. Furthermore, a visual semantic-driven camera-LiDAR fusion ranging algorithm is developed. Through rigorous spatiotemporal synchronization and cascaded outlier filtering, the integrated system achieves millimeter-level 3D localization within the typical 1.0–2.0 m operating range of agricultural robots. It maintains a Mean Absolute Error (MAE) of only 1.45 mm in the longitudinal direction at a stable processing throughput of 20 FPS. Compared to traditional pure vision depth estimation, this heterogeneous fusion approach achieves a remarkable 96.3% reduction in spatial positioning error at extended distances, fundamentally eliminating depth degradation caused by complex illumination. Ultimately, this system provides a highly robust, full-cycle geometric perception framework for the autonomous management of open-field green cabbage. Full article

A photo-curable silicone-modified resin system based on polydimethylsiloxane (PDMS) was developed and systematically evaluated for stereolithography (SLA)-based 3D printing applications. The resin formulation consisted of bisphenol A ethoxylate dimethacrylate (Bis-EMA) and trimethylolpropane triacrylate (TMPTMA) as reactive monomers, with methacrylate-terminated PDMS (PDMS-MMA) incorporated at concentrations ranging from 0 to 15 wt%. The influence of PDMS-MMA content on key physicochemical properties relevant to SLA processing, including viscosity, mechanical performance, thermal stability, optical transmittance, and curing shrinkage, was systematically investigated. Moderate incorporation of PDMS-MMA improved the mechanical flexibility of the resin, with the tensile strength reaching a maximum value of 5.95 MPa at 5 wt% PDMS-MMA. However, further increases in PDMS-MMA content resulted in a gradual decrease in tensile strength and optical transmittance, indicating the importance of optimizing the formulation composition. Thermogravimetric analysis (TGA) indicated improved thermal stability with increasing PDMS-MMA content, while curing shrinkage decreased progressively as the PDMS fraction increased. Structural printing tests confirmed that the developed resin system exhibited stable layer adhesion and shape fidelity during SLA fabrication, enabling the successful printing of complex three-dimensional structures. These results demonstrate that PDMS-modified acrylate resins provide a promising strategy for balancing mechanical flexibility, dimensional stability, and printability in SLA-based additive manufacturing. Full article

Background/Objectives: The medical mushroom Ophiocordyceps sinensis (Caterpillar Fungus), known for its ability to enhance “vitality,” is one of the most popular medicines in Asian traditional medical systems. According to the Chinese Pharmacopeia, O. sinensis is standardized for its adenosine content, the precursor of ATP, which mediates numerous physiological and pathological processes in many diseases. The related fungus of order Hypocreales, Cordyceps militaris, and its major bioactive constituents, 3′-deoxyadenosine (cordycepin), also exhibit pleiotropic biological activities. This review aims to provide a rationale for the adaptogenic and resilience-supporting effects of these medicinal fungi and to align food and drug regulation in Western countries. Methods: In this narrative review, we integrated results from chemical, pharmacokinetic, network pharmacology, preclinical, and clinical studies of O. sinensis, C. militaris, and cordycepin using network pharmacology and bioinformatics tools. Results: Across studies, recurrent mechanistic hubs included PI3K–Akt, AMPK–mTOR, MAPK, NF-κB, apoptosis, and adaptive stress-response signaling pathways, linking immune regulation and metabolic homeostasis. Experimental studies confirmed modulation of cytokine production, kinase signaling, and mitochondrial regulators. Clinical meta-analyses demonstrate consistent adjunctive benefits in renal and pulmonary disorders, although heterogeneity in preparation and methodological limitations remains significant. The review reveals controversy regarding the bioavailability of cordycepin in vivo and its concentration in vitro studies, raising the hypothesis that cordycepin may act as a driver, triggering the organism’s adaptive stress response in stress-induced and aging-related diseases. Pharmacokinetic data indicate that systemic cordycepin concentrations after oral administration remain in the nanomolar range, suggesting that some predicted molecular interactions may occur indirectly or through systems-level mechanisms. The review, for the first time, suggests establishing a regulatory category for resilience-supporting physiological modulators to align food and drug regulation in the EU with contemporary systems biology, thereby complementing the work of EFSA, EMA, FDA, and Asian authorities. Conclusions:O. sinensis, C. militaris, and 3-deoxyadenosine share a common adaptogenic mechanism for maintaining homeostasis of cellular and integrated biological system functions. The systems-level network analysis and reductionistic molecular ligand preceptor pharmacology provide complementary approaches for understanding the multi-target bioactivity of these fungi. This review clarifies conceptual and regulatory barriers to recognizing resilience-supporting interventions and informs future regulatory innovation. Full article

Underwater micro-bubble detection entails multiple challenges, including diminutive target sizes, sparse pixel information, pronounced specular highlights and water scattering, indistinct bubble boundaries, and adhesion or overlap between instances. To address these issues, we propose HSMD-YOLO, an improved detector tailored for high-resolution micro-bubble detection and built upon YOLOv11. The model incorporates three novel components: the Scale Switch Block (SSB), a scale-transformation module that suppresses artifacts and background noise, thereby stabilizing edges in thin-walled bubble regions and enhancing sensitivity to geometric contours; the Global Local Refine Block (GLRB), which achieves efficient global relationship modeling with an asymptotic linear complexity ($O\left(N\right)$) in spatial dimensions while further refining local features, thereby strengthening boundary perception and improving bubble–background separability; and the Bidirectional Exponential Moving Attention Fusion (BEMAF), which accommodates the multi-scale nature of bubbles by employing a parallel multi-kernel architecture to extract spatial features across scales, coupled with a multi-stage EMA based attention mechanism to enhance detection robustness under weak boundaries and complex backgrounds. Experiments conducted on an Side-Illuminated Light Field Bubble Database (SILB-DB) and a public gas–liquid two-phase flow dataset (GTFD) demonstrate that HSMD-YOLO achieves mAP@50 scores of 0.911 and 0.854, respectively, surpassing mainstream detection methods. Ablation studies indicate that SSB, GLRB, and BEMAF contribute performance gains of 1.3%, 2.0%, and 0.4%, respectively, thereby corroborating the effectiveness of each module for micro-scale object detection. Full article

Deep multimodal learning has excelled in tasks involving vision, language, and audio modalities. Nevertheless, their performance on tail classes exhibits significant degradation under the long-tailed distributions common in real-world data, meanwhile related fusion schemes often provide only limited treatment of modality-specific uncertainty and rarely incorporate explicit mechanisms for class-level fairness. To address these information discrepancies, we present a framework that integrates evidential reasoning with deep learning–Uncertainty-Quantified Multimodal Learning for Long-Tailed Classification (UMuLT). The framework includes: (i) an uncertainty-gated evidential fusion module that adaptively down-weights unreliable modalities; (ii) an exponential moving average (EMA) fairness regularizer that dynamically amplifies tail-class gradients; and (iii) a cross-modal consistency regularizer optimized in two stages: tail specialization with lightweight adapters on tail-class data to obtain a balanced initialization, followed by end-to-end fine-tuning. The effectiveness and practicality of our method are verified on three long-tailed benchmarks for multimodal classification. Experiments show consistent gains over strong baselines in overall metrics, calibration, and tail subset performance. Statistical significance tests confirm the superiority of the proposed framework. Full article

In the field of forensic document examination, accurately determining the chronological sequence of intersecting lines between seal ink and handwriting is a crucial technical step for verifying document authenticity, identifying contract tampering, and detecting forged signatures. This technique analyzes the physical superimposition relationship formed by the deposition of the two media on the paper substrate to provide objective scientific evidence for judicial practice. Although traditional methods such as microscopic imaging and mass spectrometry analysis have achieved some progress, they still suffer from common limitations including high equipment costs, complex operation, and potential damage to samples. This study proposes and validates an innovative non-destructive determination method that integrates structured light 3D reconstruction technology with deep learning algorithms. The research captures the microscopic 3D morphological features of the ink intersection area using a high-precision structured light scanning system and effectively eliminates noise interference caused by paper substrate undulation through Gaussian flattening technology. Subsequently, a multimodal fusion strategy combines 2D texture images with 3D depth information to construct a dataset rich in features. On this basis, a deep learning model based on an improved Residual Neural Network (ResNet) is designed, incorporating the ELU activation function and an EMA mechanism to enhance the model’s feature extraction capability and convergence stability. Experimental results demonstrate that the proposed method achieves a recognition accuracy of 94.39% on the test set, fully validating its effectiveness and application potential in the non-destructive determination of ink stroke sequencing. Full article

Background: Artificial Intelligence as a Medical Device (AIaMD) introduces regulatory, methodological, ethical, and clinical challenges that are not fully addressed by traditional device trial frameworks. Given rapidly evolving and jurisdiction-specific guidance, a consolidated mapping of trial design expectations and regulatory requirements is needed. Objective: To map regulatory requirements and clinical trial design approaches for AIaMD across major jurisdictions and to identify key methodological and implementation gaps relevant to adaptive/continuously learning systems. Methods: A scoping review was conducted in accordance with the PRISMA-ScR reporting guideline. Peer-reviewed literature (2017–2025) was searched in PubMed, Embase, Web of Science, and the Cochrane Library. Gray literature was identified from major regulators and policy bodies (FDA, EMA, MHRA, PMDA, WHO, CDSCO). Eligible records addressed AIaMD clinical evaluation, trial design, regulatory pathways, post-market surveillance, or reporting standards. Data were charted using a predefined extraction framework and synthesized descriptively with thematic analysis across regulatory, methodological, ethical, and clinical implementation domains. Results: Included sources demonstrate substantial heterogeneity in evidence expectations and AI-specific pathways across jurisdictions. Recurrent themes include the need for predefined change management, performance monitoring and drift controls, dataset representativeness and bias evaluation, transparency and versioning, cybersecurity, and real-world evidence integration. Reporting frameworks (SPIRIT-AI, CONSORT-AI, MI-CLAIM) are frequently cited as mechanisms to improve reproducibility and regulatory readiness. Conclusions: Evidence and regulatory expectations for AIaMD remain fragmented. Harmonization of terminology, trial design principles, and post-market governance—supported by standardized reporting—would improve clinical validity, safety assurance, and scalability across regions. This review has several limitations. As a scoping synthesis, it prioritizes breadth of coverage rather than quantitative meta-analysis. Included sources vary in methodological rigor and reporting detail, and evolving regulatory guidance may change rapidly over time. Nevertheless, integrating peer-reviewed and regulatory evidence provides a comprehensive overview of current expectations and emerging gaps. In conclusion, effective evaluation of AIaMD requires a shift from static, one-time validation toward continuous lifecycle oversight that integrates adaptive trial designs, transparent reporting standards, bias surveillance, and structured post-market monitoring. Regulatory heterogeneity currently poses significant barriers to multinational development; however, coordinated adoption of standardized evidence frameworks and collaborative governance mechanisms may reduce duplication while preserving patient safety. By translating methodological principles into operational guidance, this review aims to support regulators, sponsors, and clinical investigators in designing trials that are both scientifically rigorous and practically implementable for continuously learning systems. Full article

The printed circuit board (PCB), a core component of electronic products, is playing an increasingly critical role in quality defect detection. Traditional methods suffer from low efficiency and high missed detection rates, rendering them insufficient to meet the industrial requirements for PCB defect detection. To address this issue, this paper proposes an ESI-YOLOv11n model for PCB defect detection that incorporates multi-scale feature fusion. The specific improvements are as follows: First, Spatial and Channel Reconstruction Convolution (ScConv) is incorporated to optimize the C3k2 module, creating a dynamic adaptive feature extraction unit that strengthens its ability to capture features of small defects. Second, an Efficient Multi-Scale Attention (EMA) mechanism is integrated into the Neck layer, dynamically adjusting the weight distribution of multi-scale feature maps to enhance efficiency of feature fusion and improve detection performance. Finally, the Inner concept is integrated with the CIoU loss function, resulting in the novel Inner-CIoU loss function. This loss function optimizes the model by utilizing auxiliary box mechanisms and geometric constraints, leading to more accurate regression results. Experimental results show that the improved model achieves an average precision of 95.9% and a recall rate of 93.3%, which are 9.3% and 11.5% higher than those of the original model, respectively, while having a parameter size of only 13.3 Mb. The model effectively reduces the missed detection rate and false detection rate, significantly enhances the PCB defect detection performance, and demonstrates superior comprehensive performance compared with current mainstream detection models. Full article

Extracellular vesicles (EVs) have emerged as essential mediators of intercellular communication, transporting a complex repertoire of lipids, proteins, and nucleic acids that mirror the physiological and pathological status of their parent cells. This review provides a comprehensive overview of EVs from their biogenesis and molecular composition to their translational potential in human disease. This review outlines the major classes of EVs, including exosomes, microvesicles, apoptotic bodies, and oncosomes, together with recent developments in their isolation, molecular characterization, and omics-based profiling. Special focus is given to the role of EVs in viral infection, inflammation, and immune regulation, as well as their contribution to disease development and cancer biology. Moreover, we highlight the emerging clinical applications of mesenchymal stem cell-derived EVs (MSC-EVs) in regenerative medicine and oncology, alongside the therapeutic modulation of EV signaling by photobiomodulation (PBM). Finally, we address key translational challenges related to standardization, scalability, and regulatory validation. As exosome-based therapeutics fall under strict FDA and EMA oversight, their translation further depends on harmonized quality controls and robust safety evaluation. By integrating molecular mechanisms with clinical applications, this review emphasizes the transformative potential of EVs as next-generation diagnostic and therapeutic tools in precision medicine. Full article

Existing DEtection TRansformer-based (DETR) object detection methods have been widely applied to standard object detection tasks, but still face numerous challenges in detecting small objects. These methods frequently miss the fine details of small objects and fail to preserve global context, particularly under scale variation or occlusion. The resulting feature maps lack sufficient spatial and structural information. Moreover, some DETR-based models specifically designed for small object detection often have poor generalization capabilities and are difficult to adapt to datasets with diverse object scales and complex backgrounds. To address these issues, this paper proposes a novel object detection model—small object detection with efficient multi-scale collaborative attention and depth feature fusion based on DETR (ED-DETR)—which consists of three core modules: an efficient multi-scale collaborative attention mechanism (EMCA), DepthPro, a zero-shot metric monocular depth estimation model, and an adaptive feature fusion module for depth maps and feature maps. Specifically, EMCA extends the single-space attention mechanism in efficient multi-scale attention (EMA) to a composite structure of parallel spatial and channel attention, enhancing ED-DETR’s ability to express features collaboratively in both spatial and channel dimensions. DepthPro generates depth maps to extract depth information. The adaptive feature fusion module integrates depth information with RGB visual features, improving ED-DETR’s ability to perceive object position, scale, and occlusion. The experimental results show that ED-DETR achieves the current best 33.6% mAP on the AI-TOD-V2 dataset, which predominantly contains tiny objects, outperforming previous CNN-based and DETR-based methods, and shows excellent generalization performance on the VisDrone and COCO datasets. Full article

In underwater detection tasks involving bleached corals, band disease corals, and white pox disease corals, several challenges persist, including high morphological variability, difficulty in identifying small pathological regions, interference from complex underwater environments, and constraints imposed by underwater hardware. To address these issues, a lightweight underwater diseased coral target detection method, termed CD-YOLO, is proposed. Specifically, (1) a lightweight network named CDShuffleNet is constructed to replace the YOLO11 backbone, aiming to reduce model complexity while preserving detection performance; (2) a SPDConv downsampling convolution module is introduced to reduce the loss of fine-grained coral detail information during the downsampling process; and (3) attention mechanisms are incorporated through an engineering-oriented integration of EMA into the C2PSA module and the adoption of SENetV2, in order to enhance the representation of color and shape features of pathological regions and suppress interference from complex underwater environments. Experimental results demonstrate that the proposed improvements yield consistent gains in both model lightweighting and detection performance under the adopted evaluation settings. Specifically, the number of parameters, computational cost, and model size are reduced by 20.6%, 21.9%, and 18.9%, respectively, while mAP increases by 4.3 percentage points. Comparative experiments further show that the proposed method achieves a markedly higher mAP than several other state-of-the-art models. In addition, experiments conducted on the BHD Coral dataset provide preliminary evidence of cross-dataset adaptability of the proposed model. Overall, this study presents a task-oriented and application-driven improvement, demonstrating that the effective integration of lightweight components can achieve a favorable balance between model efficiency and detection performance in underwater diseased coral detection tasks. Full article

Industrial Control Systems (ICSs), critical to infrastructure, face escalating cyber threats under Industry 4.0, yet existing intrusion detection methods are hindered by attack sample scarcity, spatiotemporal heterogeneity of industrial protocols, and resource constraints of embedded devices. This paper proposes a four-stage closed-loop intrusion detection framework for ICSs, with its core innovations integrating the following key components: First, a protocol-conditioned Conditional Generative Adversarial Network (CTGAN) is designed to synthesize realistic attack traffic by enforcing industrial protocol constraints and validating syntax through dual-path discriminators, ensuring generated traffic adheres to protocol specifications. Second, a three-tiered sliding window encoder transforms raw network flows into structured RGB images, capturing protocol syntax, device states, and temporal autocorrelation to enable multiresolution spatiotemporal analysis. Third, an Efficient Multiscale Attention Visual State Space Model (EMA-VSSM) is developed by integrating gate-enhanced state-space layers with multiscale attention mechanisms and contrastive learning, enhancing threat detection through improved long-range dependency modeling and spatial–temporal correlation capture. Finally, a lightweight EMA-VSSM student model, developed via hierarchical distillation, achieves a model compression rate of 64.8% and an inference efficiency enhancement of approximately 30% relative to the original model. Experimental results on a real-world ICS dataset demonstrate that this lightweight model attains an accuracy of 98.20% with a False Negative Rate (FNR) of 0.0316, outperforming state-of-the-art baseline methods such as XGBoost and Swin Transformer. By effectively balancing protocol compliance, multi-resolution feature extraction, and computational efficiency, this framework enables real-time deployment on resource-constrained ICS controllers. Full article

With the advancement of cotton mechanized topping technology, deep learning-based methods for detecting cotton apical buds have made significant progress in improving detection accuracy. However, existing algorithms generally suffer from complex structures, large parameter counts, and high computational costs, making them difficult to deploy in practical field environments. To address this, this paper proposes a lightweight YOLO-SEW algorithm for detecting cotton apical buds in complex cotton field environments. Based on the YOLOv8 framework, the algorithm introduces Spatial and Channel Reconstruction Convolutions (SCConv) into the C2f module of the backbone network to reduce feature redundancy; embeds an Efficient Multi-scale Attention (EMA) module in the neck network to enhance feature extraction capabilities; and replaces the bounding box loss function with a dynamic non-monotonic focusing mechanism, WIoU, to accelerate model convergence. Experimental results on cotton apical bud data collected in complex field environments show that, compared to the original YOLOv8n algorithm, the YOLO-SEW algorithm reduces parameter count by 40.63%, computational load by 25%, and model size by 33.87%, while improving precision, recall, and mean average precision (mAP) by 1.2%, 2.5%, and 1.4%, respectively. Deployed on a Jetson Orin NX edge computing device and accelerated with TensorRT, the algorithm achieves a detection speed of 48 frames per second, effectively supporting real-time recognition of cotton apical buds and mechanized topping operations. Full article

Non-Intrusive Load Monitoring is an effective method for disaggregating the power consumption of individual appliances from the aggregate load data of a building. The advent of smart meters, Internet of Things devices, and artificial intelligence technologies has significantly advanced the capabilities of non-intrusive load monitoring. However, challenges such as varying sampling frequencies and measurement sensitivities remain. This paper introduces an innovative model incorporating an Adaptive Multi-Scale Attention Integration Module (AMSAIM) to address these issues. The model leverages deep learning and attention mechanisms to improve the accuracy and real-time performance of non-intrusive load monitoring. Validated on the standard UK-DALE dataset, the model consistently demonstrated superior performance. In seen scenarios, our model achieved average F1-scores approximating 0.94 and notably reduced Mean Absolute Error (MAE) values. For washing machines, it achieved an F1-score of 0.99 and MAE of 41.64, outperforming the next best method’s F1-score by 1 percentage point. In challenging unseen scenarios, the model showcased strong generalization, achieving an F1-score of 0.91 for washing machines and reducing MAE to 7.66. Furthermore, an ablation study rigorously confirmed the necessity of the AMSAIM module, showing that the synergistic integration of the efficient multi-scale attention (EMA) and the selective kernel (SK) adaptive receptive field unit is crucial for enhancing model robustness and generalization. Our results highlight the model’s potential for enhancing energy efficiency and providing actionable insights for energy management across various conditions. Full article

Lefamulin, a new, first-in-class pleuromutilin antibiotic, was recently approved by the Food and Drug Administration (FDA) and European Medicines Agency (EMA) for the treatment of patients with community-acquired bacterial pneumonia (CABP). In this context, this review aimed to evaluate its activity against the most common pathogens causing this infection. A thorough search was performed in five databases (Embase, Scopus, Web of Science, PubMed, PubMed Central) from their inception to 14th of October 2025. Clinical and Laboratory Standards Institute (CLSI) and European Committee on Antimicrobial Susceptibility Testing (EUCAST) resistance breakpoints were applied. Out of a total of 224 articles identified, 11 were deemed eligible for inclusion. Resistance among Streptococcus pneumoniae, Haemophilus influenzae, and Staphylococcus aureus isolates was 0–2.6%, 0–2.4%, and 0–4.3%, respectively. Even among isolates with specific mechanisms of resistance, such as β-lactamase-producing H. influenzae and methicillin-resistant S. aureus, resistance was below 2.4% and 3.4%, respectively. Among isolates for which no breakpoints were available (Moraxella catarrhalis, atypical pathogens, Enterococcus spp., Streptococcus spp., Haemophilus spp., and Staphylococcus spp.), MIC₉₀ values were low. An exception were isolates belonging to Enterococcus spp., which displayed MIC₉₀ values ranging from 0.25 to >16 mg/L in the two studies with relevant data. Lefamulin demonstrated broad in vitro activity against key pathogens causing CABP, making it a considerable addition to the therapeutic options for such infections, especially in cases where first-line agents cannot be used for reasons such as allergy or previous failure. Full article