Search Results (3,078)

To address challenges such as severe occlusion caused by the dense growth of blueberry fruits in natural environments, complex backgrounds, and the limited computational resources of agricultural edge devices, this study proposes BR-DETR-Prune, a lightweight object detection model oriented towards edge computing environments. Based on the RT-DETR architecture, the model introduces a PConv-based FasterNet as the backbone network, which effectively reduces memory access latency and floating-point operation costs. Furthermore, it utilizes a “Gather-and-Distribute” (GD) mechanism to reconstruct the feature fusion neck. Through the unified aggregation and multi-branch distribution of global information, it significantly enhances the model’s feature extraction capability for dense and overlapping targets. An AIFI-RepBN encoder is designed, integrating re-parameterization technology into the attention module to further reduce computational redundancy. For lightweight processing, a random channel pruning strategy based on the “Lottery Ticket Hypothesis” is adopted to perform structural compression and fine-tuning on the model, achieving a significant reduction in the number of parameters while inversely improving accuracy. The experimental results demonstrate that BR-DETR-Prune achieves an mAP@0.5 of 97.1% on a self-built blueberry dataset, with only 15.52 M parameters and a computational load reduced to 34.0 GFLOPs. Its comprehensive performance is superior to mainstream models such as YOLOv8, YOLO11, and the original RT-DETR. Particularly, deployment testing on the NVIDIA Jetson Orin Nano Super embedded edge computing platform reveals that the model achieves a real-time inference speed of 20.5 FPS under FP16 precision, exhibiting smooth detection frames and strong robustness against occlusion. This study provides an effective optimization solution for the deployment of high-precision Transformer architectures on low-computational-power devices, offering an efficient and reliable visual perception approach for automated blueberry harvesting and yield estimation. Full article

The growing sophistication of cyber threats has reduced the effectiveness of traditional cybersecurity tools in protecting modern organisations and complex networks. This challenge requires advanced solutions capable of real-time detection, rapid response, and efficient threat mitigation. In this context, AI-based approaches have emerged as a powerful enabler of intelligent, adaptive, and data-driven security operations. This study presents a comprehensive analysis of AI-driven threat detection combined with automated incident response mechanisms in modern cybersecurity architectures. The novelty of this work lies in the integration of advanced machine learning-based detection with real-time, automated response capabilities to address zero-day and previously unknown threats in heterogeneous digital environments. The paper examines system architecture design, implementation strategies, and performance evaluation across diverse deployment scenarios. Experimental results demonstrate that AI-driven detection with automated response significantly enhances cybersecurity effectiveness, achieving accuracies between 96% and 97%, dramatically reducing the mean response time from 45 min to less than 30 s, and substantially improving zero-day threat detection and containment success rates. Overall, the proposed approach achieves up to a 98.9% improvement in incident containment efficiency, highlighting the operational and defensive advantages of intelligent automation. Full article

Software aging and the corresponding need for system rejuvenation are well-established concepts in computer science. As virtualization technologies are increasingly adopted within electric power utility infrastructures, early investigation into Software Aging and Rejuvenation (SAR) models, aging indicators, and empirical data collection becomes essential. Given the critical role of the electric power grid and the high dependability requirements of the protection and control systems that support its operation, proactive research in this area is timely and necessary. Motivated by this need, this work proposes a hierarchical framework that integrates an SAR model into the Reliability Block Diagram (RBD) representation of a Digital Substation Automation System (DSAS). The analysis shows that, for the selected parameter set, incorporating SAR into the VPAC reliability model results in higher estimated failure rates and increased annual downtime relative to hardware-only models. When combined with substation primary system indices, however, the overall reliability indices remain largely unchanged, aside from reduced outage duration attributed to improved switching performance enabled by the DSAS architecture. Further examination reveals that the limited influence of SAR is primarily due to the lack of historical failure-mode data for the secondary system. Availability of such empirical data is expected to significantly affect combined reliability indices and improve the accuracy of reliability evaluations. This highlights the importance of systematic data collection and aging-indicator analysis as utility infrastructures transition toward virtualized and software-dependent architectures. Full article

Critical infrastructure restoration (CIR) is a disaster-management and sustainability challenge because prolonged disruption of energy, water, transport, communications, healthcare, and public-administration services can amplify social, economic, and environmental losses. This PRISMA 2020-reported systematic review synthesizes post-2016 scientific literature and official policy, legal, standards, and technical documents on CIR and AI decision support. The review identified 55 records, removed 1 duplicate, excluded 1 ineligible record, and retained 53 core sources for qualitative synthesis, including 31 scholarly publications and 22 official documents. Manual screening was used; no automated screening or AI-assisted exclusion tools were applied. The results are organized around four research questions covering regulatory frameworks, recovery practices, supporting systems, and AI model families. The synthesis shows that CIR is shaped by layered governance through NIS2, the CER Directive, the AI Act, and national measures; by operational recovery practices such as continuity planning, cyber crisis coordination, interdependency mapping, and model-supported restoration; by digital platforms including SCADA/ICS, IoT sensing, GIS/common operating pictures, decision-support systems, simulation environments, and digital twins; and by AI methods ranging from classical machine learning and computer vision to reinforcement learning and generative assistants. However, evidence maturity remains uneven, with many AI applications still simulation-based, sector-specific, or weakly validated in real restoration settings. The review contributes an integrated CIR-oriented framework showing that AI creates practical value when embedded in interoperable, human-supervised, regulation-aware, and empirically validated restoration architectures that support sustainable service continuity rather than isolated automation. Full article

Accurate and automatic segmentation of the liver and liver tumors from computed tomography (CT) images is essential for computer-assisted diagnosis, treatment planning, and clinical decision-making. Although deep learning-based segmentation models, particularly U-Net and its variants, have achieved promising results in medical image analysis, many existing approaches mainly focus on local pixel-level feature extraction and may have limited ability to explicitly model long-range spatial relationships among anatomically meaningful regions. In addition, liver tumor segmentation remains challenging due to low contrast, irregular tumor boundaries, heterogeneous tumor appearances, and noise or artifacts in CT images. To address these limitations, this study proposes a hybrid ensemble neural network architecture that integrates an improved U-Net and a Graph U-Net for automatic liver and liver tumor segmentation. The improved U-Net is designed to capture fine-grained local features and preserve detailed spatial information through an encoder–decoder structure with skip connections, while the Graph U-Net uses Simple Linear Iterative Clustering (SLIC)-based superpixels to construct a graph representation of CT images and model spatial dependencies between adjacent image regions. By combining these complementary representations through an ensemble learning strategy, the proposed framework enhances both pixel-level segmentation accuracy and robustness against noisy imaging conditions. The proposed method was evaluated on the LiTS17 dataset, where CT images were preprocessed using intensity filtering, resizing, data augmentation, and normalization. Experimental results demonstrate that the proposed ensemble architecture achieves 99.2% accuracy for liver segmentation and 98.1% accuracy for liver tumor segmentation, outperforming representative segmentation models such as MultiresUnet and R2U-Net. Furthermore, robustness experiments under different signal-to-noise ratio conditions show that the proposed model maintains stable performance in noisy CT images, achieving 85% accuracy even under severe noise at −4 dB SNR. This result highlights the advantage of integrating convolutional feature learning with graph-based spatial relationship modeling for improving segmentation stability when image quality is degraded by noise or artifacts. These findings indicate that the integration of improved U-Net, SLIC-based graph construction, and Graph U-Net provides an effective and noise-robust solution for liver and liver tumor segmentation, with potential applicability as a computer-assisted tool in clinical image analysis after further validation on larger and external datasets. Full article

Misinformation has become a persistent feature of contemporary digital information environments. Platform designs and business models often privilege attention, engagement, and repeated exposure over epistemic quality. However, misinformation does not diffuse uniformly across platform structures. This study examines how contested claims in a South Korean social policy controversy circulate on YouTube. The analysis focuses on unfounded allegations regarding permanent employment offers to part-time workers at Incheon International Airport across two analytic levels: (1) a videoclip network, in which video-to-video ties are formed through shared commenters over time, and (2) a channel network, in which channel-to-channel ties are formed through shared commenters over time. Drawing on YouTube Data API records, we employ a mixed computational approach that integrates social network analysis, speech-to-text transcription, natural language processing, semantic network analysis, and automated content classification. Videos are classified as misinformation or non-misinformation based on the presence of demonstrably incorrect claims or corrective content. We compare network structure, diffusion patterns, and engagement dynamics across these two layers. The results reveal pronounced layer-specific differences. Misinformation diffuses more extensively within the channel network, which exhibits higher density and stronger cross-channel interconnectedness, suggesting that creator-level infrastructures function as stabilising conduits for the circulation of false claims. By contrast, diffusion pathways at the videoclip level show comparatively weaker differentiation between misinformation and non-misinformation content. Engagement patterns also diverge misinformation videos attract significantly more likes, while message format and channel attributes are less consistently distinguishing. From a theoretical standpoint, this study advances a multi-layer diffusion perspective on platform-mediated misinformation by demonstrating how platform architectures shape the visibility, persistence, and amplification of false claims. The findings highlight the importance of intervention strategies that move beyond individual content moderation toward creator- and network-level governance mechanisms, with implications for the design of platform features, recommendation systems, and misinformation mitigation tools. Full article

Cybersecurity testing laboratories must produce auditable conformity evidence while operating with rapidly changing toolchains, conditional requirements, and qualitative PASS/FAIL/INCONCLUSIVE outcomes. ISO/IEC 17025:2017 is widely used to demonstrate laboratory competence, yet its operationalisation in cybersecurity testing remains under-specified for software- and tool-driven security assessments. This paper separates an architectural contribution from an empirical contribution. The architectural contribution is a digitalized quality-management framework and automation-ready compliance architecture that translate ISO/IEC 17025 clauses into cybersecurity-specific artefacts, decision rules, controlled toolchains, evidence bundles, and review workflows. The empirical contribution is an exploratory single-laboratory case study based on unpublished, anonymised, and confidentiality-constrained laboratory artefacts: an ETSI TS 103 701 workbook with 68 provision-level test groups, including 41 claimed/applicable rows for ambiguity analysis; an IEC 62443 corrective-action plan; and ISO/IEC 17025 governance records. Within this case, structured decision rules and evidence traceability reduced the Conformity Statement Ambiguity Index from 0.976 to 0.049 and converted 37 previously INCONCLUSIVE provisions into PASS determinations. These results are reported as descriptive within-case evidence only; they do not establish predictive validity or cross-laboratory generalisability. The study contributes a clause-to-artefact crosswalk, a concrete evidence-traceability architecture, and candidate cyber-maintenance indicators for future multi-laboratory validation. Full article

With the acceleration of urbanization and the intensification of climate change, wind conditions have become a critical factor in architectural design. They not only affect a building’s wind resistance but also influence ventilation, pollutant dispersion, pedestrian comfort, and energy consumption. Traditional computational fluid dynamics (CFD) simulations are costly. Although the application of machine learning for CFD prediction has become a relatively mature technology, machine learning models still face challenges in actual architectural design workflows. Building upon recent advancements in the field, it proposes two core technologies: a method for predicting outdoor wind environments in buildings based on the Site-Specific Training for Design Tasks (SST-DT) strategy, and an automated machine learning workflow. These innovations improve upon existing wind environment analysis methods and systems, establishing a fully automated working framework that is easy for architects to learn and use. Within this framework, dataset acquisition and model training are performed automatically. Finally, this framework was validated across various prediction tasks with different objectives. It significantly lowers the barrier to entry for architects adopting machine learning, advances the performance-driven design paradigm, and facilitates the deep integration of machine learning technologies into architectural wind engineering. Full article

Background: Cardiac magnetic resonance (MRI) images often exhibit low contrast, anatomical variability, and indistinct boundaries, particularly in the myocardium (MYO) and right ventricle (RV). These challenges can reduce the reliability of both manual and automated segmentation, highlighting the need for more robust and boundary-aware approaches. Methods: In this study, an Aspect-Aware Boundary-Resilient UNet3D (ABR-UNet3D) architecture is proposed for cardiac MRI segmentation. The model incorporates an Aspect-Aware Complementary Attention (AAC) module that combines multi-planar contextual information with a complementary gating mechanism to enhance boundary representation. The method was evaluated on the ACDC dataset under consistent training conditions. In addition to Dice Similarity Coefficient (DSC) and Intersection over Union (IoU), boundary-based metrics, including the 95th percentile Hausdorff Distance (HD95), Average Surface Distance (ASD), and Surface Dice, were employed. Furthermore, a five-fold cross-validation protocol and detailed ablation studies were conducted to assess robustness and analyze the contribution of individual AAC components. Results: The proposed method achieved a mean DSC of 0.9603 in single-run experiments on the ACDC dataset and showed consistent performance in anatomically challenging regions, particularly for RV and MYO segmentation. In addition, five-fold cross-validation experiments resulted in an average DSC of 0.952 ± 0.009 and IoU of 0.908 ± 0.012, indicating stable performance across different data splits within the evaluated dataset. Boundary-based metrics also showed improved surface agreement and lower boundary errors compared with the evaluated baseline models. Ablation studies further indicated that the combined use of multi-planar contextual information and complementary gating contributes more effectively to segmentation performance than the individual components used separately. Conclusions: The results suggest that the proposed ABR-UNet3D architecture provides a stable and competitive segmentation framework for cardiac MRI images within the scope of the ACDC dataset. By jointly modeling contextual information and boundary refinement, the method improves segmentation reliability in challenging regions while maintaining competitive and consistent performance with respect to existing approaches. Full article

Detecting the presence of carbonation is critical for monitoring structural safety and durability. Identifying the presence of carbonation reveals the risk of chemical changes within the concrete and the potential for reinforcement corrosion. This detection allows for a reliable and prioritized assessment of the structure’s current condition. Therefore, checking for the presence or absence of carbonation is a critical indicator in determining structural safety and maintenance priorities. This study explicitly addresses a critical gap in the literature, where existing carbonation research predominantly focuses on regression-based estimation of carbonation depth, while the problem of direct visual classification of carbonation presence for rapid decision-making currently remains underexplored. In this context, the study aims to fill this research gap through developing a robust and field-applicable deep learning-based classification framework for the automated detection of carbonation presence on concrete surfaces using images, while systematically comparing the performance of different YOLO architectures and assessing the suitability of a previously unused dataset (ConcreteCARB) for carbonation classification tasks. In this context, YOLOv8m, YOLOv11m, YOLOv12m, and YOLOv26m were compared for concrete carbonation classification, aiming to find the most suitable model. The results show that YOLOv8m and YOLOv11m achieve perfect accuracy (Accuracy = 0.9981, Precision = 1, Recall = 0.9964, Specificity = 1, AUC-ROC = 1). In inference efficiency analyses, the YOLOv11m model was identified as the fastest model with the lowest latency and highest FPS. While YOLOv8m and YOLOv26m offered balanced speed-performance results, YOLOv12m showed a relatively lower processing speed. The findings indicate that YOLOv11m is the most suitable option for real-time applications. Full article

Capacitive sensing offers a low-power, non-optical route for automated insect monitoring, but architecture-level benchmarking under shared geometry remains limited. Rather than presenting a general framework, this study proposed a configuration-specific laboratory benchmark comparing four sigma-delta and charge-transfers in a 6 mm dual-arc conduit at 25 °C, targeting six adult terrestrial arthropod species spanning a 25-fold range of the body cross-sectional area. Static measurements showed a strong linear relationship between ΔC_static and body cross-sectional area (17.96 fF/mm², r = 0.995), supporting first-pass conduit sizing and detectability screening. In contrast, transit amplitudes were not monotonic with body size because posture, motion, and gap occupancy affected waveform shape. Under chamber conditions, static sensitivity degraded by less than 3.2% across all architectures from RH 40% to 80%. However, under the deployment-oriented noise model, SNR_FR degradation was substantially higher for charge-transfer devices (64.8–66.8%) than for Σ–Δ devices (≤35.5%), because the composite noise floor amplifies the effect of humidity-induced baseline drift. These results generated a conduit-specific reference dataset for preliminary capacitance-to-digital converter (CDC) selection within the tested 6 mm dual-arc geometry. In addition, the experimental validation focused on laboratory baseline noise characterization, long-term drift, and trap-integrated testing in temperature-controlled environments and natural-locomotion trials, providing critical information on configuration-specific architectures and body-size-scaling reference. This study serves as an initial step toward real-world capacitive insect sensing. Future studies will investigate additional conduit geometries and insect species to improve the robustness of the proposed framework. Full article

Objective: Chiari Type I Malformation (CM1) is a structural abnormality of the hindbrain that can cause a range of neurological symptoms and often requires radiological confirmation using magnetic resonance imaging (MRI). The aim of this study was to develop and evaluate a deep feature-based machine learning framework for the automated detection of CM1 from sagittal MRI images. Materials and Methods: The cohort comprised 550 adults: 250 patients with CM1 (168 women, 82 men; age range, 18–65 years) and 300 healthy control participants (210 women, 90 men; age range, 18–65 years). A total of 764 T2-weighted sagittal MR images (384 CM1, 380 healthy) acquired from two different 1.5T MRI scanners (Siemens Magnetom Altea and Symphony) between 2020 and 2024 were retrospectively analyzed. Deep features were extracted using ResNet-50 and MobileNetV2 architectures and subsequently classified using Support Vector Machines (SVM), Logistic Regression (LR), Random Forest (RF), XGBoost, and voting-based ensemble models. Model performance was assessed through patient-level 5-fold cross-validation using accuracy, sensitivity, specificity, F1-score, PPV, NPV, and AUC metrics. Code and trained models are available from the corresponding author upon reasonable request; imaging data are not publicly available due to patient privacy and institutional restrictions. Results: Across patient-level five-fold cross-validation, models built on ResNet-50 deep features demonstrated extremely high and stable diagnostic performance. The final soft-voting ensemble classifier based on ResNet-50 achieved perfect mean performance, with accuracy, balanced accuracy, sensitivity, specificity, F1-score, and AUC all equal to 1.000 ± 0.000 across folds. Other ResNet-based classifiers also achieved near-perfect results. MobileNetV2-based models also demonstrated strong performance but showed slightly lower stability compared with ResNet-based models, with mean accuracies ranging from 0.984 to 0.993 and mean AUC values between 0.99947 and 0.99984 across classifiers. Conclusion: The proposed deep feature-based machine learning framework demonstrated excellent performance for the automated detection of Chiari Type I Malformation from sagittal MRI images. In particular, the ResNet-50–based soft-voting ensemble model achieved perfect classification performance in cross-validation testing, suggesting that deep feature representations combined with machine learning classifiers may serve as a promising computer-aided diagnostic tool for supporting radiological evaluation of CM1. Full article

Automated depression assessment is critical for scalable mental healthcare but faces dual challenges: the lack of clinical interpretability in “black-box” deep learning models and the excessive computational cost of large-scale fusion architectures. To bridge this gap, we propose NS-Dep-KAN, a novel neuro-symbolic framework that harmonizes DSM-5-guided reasoning with Kolmogorov–Arnold Networks (KANs). Our approach leverages a Large Language Model (LLM) to extract symbolic symptom evidence aligned with diagnostic criteria, which then guides the aggregation of multimodal features from frozen pretrained encoders (WavLM and Qwen). Unlike traditional Multi-Layer Perceptrons, the proposed KAN prediction head employs learnable B-spline activation functions to capture complex nonlinear symptom–severity mappings with extreme parameter efficiency. Evaluations on the DAIC-WOZ benchmark demonstrate that NS-Dep-KAN achieves state-of-the-art performance among audio-text models (MAE 2.69, 13.5% improvement over the three-modality baseline MSGAF at MAE 3.11), with only ∼4.9 K trainable parameters. Moreover, the framework offers inherent interpretability, revealing granular symptom contribution profiles that align with clinical intuition. This work establishes a path toward explainable trustworthy AI for mental health screening. Full article

Shipboard diesel–electric power systems (SDEPSs) are inherently vulnerable to transient instability owing to their compact, isolated, and low-inertia design. Their performance is considerably influenced by dynamic disturbances, which can lead to operational failures and accidents of varying severity. Therefore, this research addresses the critical challenge of transient stability enhancement in SDEPSs during significant dynamic disturbances. Recognizing that traditional automation and protection systems respond only after transient instability occurs, this study introduces an emergency preventive control (EPC) strategy that enables anticipatory control of SDEPS power sources to enhance transient stability. The proposed EPC system integrates hardware and software components to perform real-time monitoring and control based on forecasting system parameters, specifically the relative rotor angles of the power sources. The feasibility and effectiveness of the proposed system are validated through comprehensive computer simulations, demonstrating improvements in transient stability and system resilience by substantially reducing relative rotor angle deviations during the transient event. Overall, the proposed framework can be readily integrated into existing shipboard control architectures, offering an effective means to improve the safety of modern SDEPSs operating under dynamic conditions. Full article

Background/Objectives: Epiretinal membrane (ERM) is a common vitreoretinal disorder characterized by fibrocellular proliferation on the inner retinal surface, often leading to progressive visual impairment. Accurate grading of ERM severity using optical coherence tomography (OCT) is critical for treatment planning and surgical decision-making; however, manual grading is labor-intensive and subjective. This study aims to develop an automated and reliable deep learning-based method for ERM severity classification. Methods: We propose DenseViT-OCT, a hybrid deep learning model that integrates dense convolutional neural networks (CNN) and vision transformers (ViT). The model introduces three key modules: Multi-Scale Dense Feature Aggregation (MDFA) for capturing hierarchical features across multiple spatial scales, Adaptive Feature Calibration (AFC) for enhancing feature discrimination through channel and spatial attention, and Cross-Attention Feature Fusion (CAFF) for enabling bidirectional interaction between convolutional and transformer representations. The model was trained and evaluated on 2195 OCT B-scan images obtained from 397 patients. Results: DenseViT-OCT achieved an overall accuracy of 94.76% on the internal four-class test set, outperforming 19 benchmark models, including ConvNeXt, EfficientNet, ViT, and Swin Transformers. The model demonstrated balanced performance with a macro-averaged precision of 93.76%, recall of 93.22%, F1-score of 93.47%, Cohen’s kappa of 92.62%, and macro-Area Under the Curve (AUC) of 98.95%. Ablation experiments confirmed the contribution of the proposed MDFA, AFC, CAFF, and deep supervision components, with the full model consistently outperforming reduced variants and standalone DenseNet121 and ViT-B/16 backbones. In repeated experiments across five random seeds, DenseViT-OCT also achieved the best mean accuracy (0.9399 ± 0.0052). External validation on the public multicenter OCTDL dataset, performed as binary ERM-versus-normal classification because of label availability, yielded 90.76% accuracy and 97.61% AUC, indicating promising generalization beyond the development cohort. Conclusions: DenseViT-OCT provides a robust framework for automated ERM severity classification from OCT B-scans. The combination of local CNN features, global transformer context, and dedicated fusion modules improves classification performance and yields clinically meaningful error patterns. Although further stage-wise multicenter validation, volumetric OCT analysis, and prospective clinical assessment are required, the proposed method shows promise as a research-oriented decision-support framework for B-scan-level ERM assessment. Full article