Search Results (3,122)

Convolutional neural networks (CNNs) excel in tasks such as image, speech, and video recognition, as well as pattern analysis. However, their reliance on large training datasets, often sourced from third-party providers, exposes them to security risks, particularly poisoning attacks. Targeted poisoning attacks, also known as backdoor attacks, enable a CNN model to correctly classify normal data while misclassifying inputs containing specific triggers. In contrast, untargeted poisoning attacks aim to degrade the overall performance of the model. This research introduces an invisible targeted poisoning attack characterized by low implementation complexity and high computational efficiency due to its computationally inexpensive LSB-based embedding mechanism, without requiring complex optimization procedures against a basic CNN model and a residual network (ResNet-18) model. By embedding trigger images within poisoned samples, the attack remains covert, evading detection. The model is then trained on a dataset comprising both original and poisoned samples. The expected outcome is that the model will classify regular images correctly, but will misclassify those containing the embedded trigger as belonging to a target class. Experimental results on the CIFAR-10 dataset demonstrate the effectiveness of this approach, achieving a 99.32% Adversarial Success Rate (ASR) against ResNet-18 with only a 0.02% reduction in accuracy on benign test samples. Full article

Purpose: The aim of this study was to develop a concept for adjustment planning of intraocular lens orientation axes after cataract surgery with implantation of toric intraocular lenses (tIOLs) and to predict the spectacle refraction after tIOL re-alignment. Methods: This calculation concept based on paraxial spherocylindrical vergence transformations uses the actual spherocylindrical refraction at the spectacle plane, corneal power, and the labelled power and measured axis of the implanted tIOL to minimise the refractive cylinder by simulating the rotation of the tIOL in the eye. The axial lens position is derived from simple prediction models using anterior chamber depth and lens thickness or axial length from preoperative biometry or the equivalent tIOL power. The new target axis is predicted together with the spherocylindrical refraction after re-alignment of the tIOL. Results: To show the applicability of this calculation model, we provide four clinical working examples: example 1 deals with keratometric power values; example 2 deals with keratometric curvature values, including surgically induced astigmatism and a statistical posterior astigmatism correction for the cornea (both examples with a thin cornea model); example 3 deals with corneal curvature data for the front and back surface; and example 4 deals with keratometric power data and corneal back surface power data, including surgically induced astigmatism (both examples with a thick cornea model). Conclusions: The effect of tIOL axis adjustment after cataract surgery can be predicted based on actual refraction, corneal power, tIOL power and the measured axis, and a simulation of the tIOL axis rotation enables the best orientation with the lowest refractive cylinder at the spectacle plane to be found. Full article

Distributed optical fiber vibration sensing (DVS) has shown strong potential in perimeter security, pipeline leakage monitoring, transportation safety, and structural health diagnostics owing to its high sensitivity, long-range coverage, and immunity to electromagnetic interference. However, severe cross-scene distribution mismatch is often encountered in real-world deployments: indoor, outdoor, and pipeline environments exhibit markedly different noise patterns and time–frequency characteristics, thereby degrading the generalization ability of models trained in a single scene. To address this challenge, we propose a Progressive Multi-Strategy Domain Alignment (PMSDA) framework for label-disjoint cross-scene vibration recognition. PMSDA uses a compact expansion–compression encoder together with complementary alignment mechanisms—maximum mean discrepancy (MMD), correlation alignment (CORAL), and adversarial domain discrimination—to learn a scene-robust latent space from a labeled indoor source and two unlabeled target domains (outdoor and pipeline) within a single alternating-training model. Because the fine-grained source and target label spaces are disjoint, PMSDA is formulated as a representation-transfer framework rather than a standard label-shared unsupervised domain adaptation method; target-domain recognition is therefore performed through domain-specific prototype clustering in the aligned latent space. On three representative scenes with nine event classes in total, PMSDA achieved 89.5% accuracy, 86.7% macro-F1, and 0.93 AUC for Indoor→Outdoor, and 85.8%, 84.7%, and 0.87, respectively, for Indoor→Pipeline, outperforming traditional feature+SVM/RF pipelines, CNN/ResNet baselines, and representation-transfer baselines adapted from DANN/CDAN/SHOT under the same evaluation protocol. These results indicate that PMSDA is a promising and effective framework for offline cross-scene DVS evaluation under disjoint target event sets. Full article

Introduction: Spondyloenchondrodysplasia with immune dysregulation (SPENCDI) is a rare autosomal recessive disorder caused by biallelic variants in the tartrate-resistant acid phosphatase 5 (ACP5) and characterized by variable skeletal, immunological, and neurological manifestations. Because early skeletal abnormalities may be subtle, diagnosis can be challenging in infancy. Materials and methods: We conducted a detailed clinical, immunological, radiological, and molecular evaluation of an infant with early-onset cytopenia, recurrent infections, seizures, and developmental delay. Genomic analysis was performed using whole exome sequencing (WES) and copy number variation sequencing (CNV-seq). In addition, we performed a structured narrative review of published ACP5-related SPENCDI cases to summarize the clinical spectrum and the currently reported use of Janus kinase (JAK) inhibitors. Results: Genomic analysis identified an ACP5 stop-gain variant (c.311G>A; p.Trp104*) with an apparently homozygous signal on WES. Re-evaluation of the copy-number data demonstrated an overlapping heterozygous 19p13.2–p13.13 deletion encompassing ACP5, indicating biallelic ACP5 defects consisting of a sequence variant on one allele and deletion of the other allele. Clinically, the patient showed prominent extra-osseous manifestations, including impaired T- and NK-cell cytotoxicity, before the emergence of definite radiographic skeletal abnormalities. Our literature review showed that skeletal abnormalities were repeatedly documented across published ACP5-related SPENCDI reports, although radiographic changes were often subtle and could be preceded by immune manifestations. Reported use of JAK inhibitors suggests potential benefit for immune dysregulation in selected patients, whereas the neurological response remains uncertain. Conclusions: This study reports a novel ACP5 variant and expands the known phenotypic spectrum of SPENCDI. SPENCDI should be considered in children with unexplained immune dysfunction and developmental delay, and suggestive neuroimaging findings, even when overt skeletal deformities are absent. Early genetic testing and targeted skeletal imaging may facilitate diagnosis. Full article

The ATRX gene encodes an SWI/SNF-type chromatin remodeling protein that is critical for proper development of the mammalian central nervous system and musculoskeletal system. While significant progress has been made in understanding the molecular functions of the full-length (FL) ATRX protein, there is still very little known about its conserved alternative spliceoform, ATRXt. ATRXt is a truncated isoform of ATRX which lacks the entire SWI/SNF domain due to the retention of intron 10, which results in the in-frame addition of 61 unique amino acids (exon 10a) at its C-terminus. Here, we demonstrate that ATRXt accounts for 5–20% of total ATRX protein levels, while showing tissue- and differentiation-specific changes in expression levels compared to its full-length counterpart. ATRXt shows enriched localization at H3K9me3-positive heterochromatin but not at PML-nuclear bodies, while physically interacting with the known FL-ATRX protein partners, DAXX and HP1α. Exon 10a can target a GFP-fusion protein to the nucleolus, but removal of exon 10a from ATRXt does not prevent nucleolar localization. Finally, re-introducing ATRXt into the ATRX-negative U2OS cell line reduced rRNA transcription and severely hampered cell growth, similar to previous studies using FL-ATRX. Our study highlights that ATRXt has many of the same properties as FL-ATRX, suggesting that some roles of ATRX do not require remodeling activity, while highlighting the need to distinguish ATRXt’s functions from those of the full-length protein. Full article

Against the global carbon neutrality target, predictive maintenance (PdM) of automotive engines represents a core technical strategy to advance the sustainable development of the automotive industry. Conventional single-modal diagnostic approaches for engine abnormal sound defects suffer from low accuracy and weak anti-interference capability. Existing multi-modal fusion methods fail to deeply mine the physical coupling between cross-modal features and often entail excessive model complexity, hindering deployment on resource-constrained on-board edge devices. To resolve these limitations, this study proposes a Physical Prior-Embedded Cross-Modal Attention (PPE-CMA) mechanism for lightweight multi-modal fusion diagnosis of engine abnormal sound defects. First, wavelet packet decomposition (WPD) and mel-frequency cepstral coefficients (MFCC) are integrated to extract time-frequency features from engine audio signals, while a channel-pruned ResNet18 is employed to extract spatial features from engine thermal imaging and vibration visualization images. Second, the PPE-CMA module is designed to adaptively assign attention weights to audio and image features by exploiting the physical coupling between engine fault acoustic and visual characteristics, enabling efficient cross-modal feature fusion with redundant information suppression. A rigorous theoretical derivation is provided to link cosine similarity with the physical correlation of engine fault acoustic-visual features, justifying the attention weight constraint (β = 1 − α) from the perspective of fault feature physical coupling. Third, an improved lightweight XGBoost classifier is constructed for fault classification, and a hybrid data augmentation strategy customized for engine multi-modal data is proposed to address the small-sample challenge in industrial applications. Ablation experiments on ResNet18 pruning ratios verify the optimal trade-off between diagnostic performance and computational efficiency, while feature distribution analysis validates the authenticity and effectiveness of the hybrid augmentation strategy. Experimental results on a self-constructed multi-modal dataset show that the proposed method achieves 98.7% diagnostic accuracy and a 98.2% F1-score, retaining 96.5% accuracy under 90 dB high-level environmental noise, with an end-to-end inference speed of 0.8 ms per sample (including preprocessing, feature extraction, and classification). Cross-engine and cross-domain validation on a 2.0T diesel engine small-sample dataset and the open-source SEMFault-2024 dataset yield average accuracies of 94.8% and 95.2%, respectively, demonstrating strong generalization. This method effectively enhances the accuracy and robustness of engine abnormal sound defect diagnosis, offering a lightweight technical solution for on-board real-time fault diagnosis and in-plant online quality inspection. By reducing engine fault-induced energy loss and spare parts waste, it further promotes energy conservation and emission reduction in the automotive industry. Quantified experimental data on fuel efficiency improvement and carbon emission reduction are provided to substantiate the ecological benefits of the proposed framework. Full article

Timely and accurate detection of flax (Linum usitatissimum L.) pests and diseases is essential for ensuring stable yield and product quality. Nevertheless, practical field deployment remains challenging due to complex backgrounds, large variations in target scale, and limited computational resources. To tackle these issues, a lightweight optimization framework based on YOLOv11 is developed for flax pest and disease recognition. A dedicated dataset comprising seven common flax pest and disease categories is established, and an instance-level data augmentation strategy is employed to enhance data diversity and mitigate class imbalance. Building upon the YOLOv11 baseline, ADown and C3K2-STAR modules are incorporated to strengthen multi-scale feature representation while reducing computational redundancy, and auxiliary detection heads are introduced to provide additional supervision during training. Experimental results show that the proposed approach achieves better performance than Faster R-CNN, RT-DETR-ResNet50, YOLOv3-tiny, YOLOv5n, YOLOv8n, YOLOv11n, and YOLOv12n by 4.4, 7.3, 4.1, 2.4, 2.6, 2.2, and 2.8 percentage points in mAP@50, respectively. For the more stringent mAP@50:95 metric, the proposed model also outperforms Faster R-CNN, RT-DETR-ResNet50, YOLOv3-tiny, YOLOv5n, YOLOv11n, and YOLOv12n by 2.9, 3.6, 4.3, 2.8, 2.0, and 0.5 percentage points, respectively, while achieving performance comparable to YOLOv8n. Meanwhile, it achieves substantial model compression, with 55.1%, 94.8%, 77.0%, 18.6%, 15.1%, and 14.5% fewer parameters than Faster R-CNN, RT-DETR-ResNet50, YOLOv3-tiny, YOLOv8n, YOLOv11n, and YOLOv12n, respectively, and it reduces computational cost (GFLOPS) by 31.5%, 95.6%, 61.5%, 19.1%, 12.7%, and 12.7%. These results indicate that the proposed method achieves a favorable balance between detection accuracy and computational efficiency, suggesting its potential suitability for practical flax pest and disease monitoring as well as deployment on resource-constrained edge devices, although further validation in more diverse scenarios is still needed. Full article

The progressive decline in functional β-cell mass in Type 2 Diabetes (T2D) is increasingly recognized not as a simple apoptotic loss, but as a complex erosion of cellular identity termed “dedifferentiation.” Central to this phenotypic shift is the metabolo-epigenetic axis, where mitochondria act as the primary sensing hub, transducing nutrient flux into biochemical signals that govern the chromatin landscape. This review synthesizes current evidence on how mitochondrial metabolites—including Acetyl-CoA, α-ketoglutarate, and NAD+—serve as obligatory co-factors for the epigenetic machinery. We explore how chronic metabolic stress triggers a “Systemic epigenetic destabilization,” leading to the loss of lineage-specific markers and the formation of persistent “metabolic scars.” Furthermore, we discuss the clinical implications of these changes, specifically regarding the phenomenon of metabolic memory and the molecular limits of β-cell reversibility. By integrating foundational transcriptional studies with emerging epigenomic data, we propose that targeting the mitochondrial–epigenetic axis offers a strategic window for re-differentiating failing β-cells and restoring glycemic homeostasis. Full article

Anti-unmanned aerial vehicle (Anti-UAV) detection is critical for airspace security, yet existing single-modality approaches suffer from severe performance degradation under adverse illumination, thermal crossover, and extreme scale variation. In this paper, we propose CSFADet, a dual-modal detection framework that jointly exploits visible and infrared imagery through four tightly integrated modules. First, a Cross-Spectral Feature Alignment (CSFA) module performs early-stage spectral calibration by computing cross-modal query–value attention maps, generating modality-aware channel descriptors that re-weight and concatenate the two spectral streams. Second, a Dual-path Texture Enhancement Module (DTEM) enriches fine-grained spatial details via cascaded convolutions with residual connections. Third, a Dual-path Cross-Attention Module (DCAM) introduces a feature-shrinking token generation strategy followed by symmetric cross-attention branches with learnable scaling factors, Squeeze-and-Excitation recalibration, and a $1\times 1$ convolution fusion head, enabling deep bidirectional interaction between modalities. Fourth, a Dual-path Information Refinement Module (DIRM) embeds Adaptive Residual Groups (ARGs) that cascade Multi-modal Spatial Attention Blocks (MSABs) with channel and dynamic spatial attention, culminating in a Multi-scale Scale-aware Fusion Refinement (MSFR) unit that employs three parallel multi-head attention branches with a Scale Reasoning Gate and Channel Fusion Layer to produce scale-discriminative enhanced features. Experiments on the public Anti-UAV300 benchmark show that CSFADet achieves 91.4% mAP@0.5 and 58.7% mAP@0.5:0.95, surpassing fifteen representative detectors spanning single-stage, two-stage, YOLO-family, and Transformer-based categories. Ablation studies confirm the complementary contributions of each module, and heatmap visualizations verify the model’s capacity to focus on small, distant UAV targets under challenging conditions. Full article

Urban air pollution is increasingly recognised as a determinant of mental wellbeing, yet most existing studies rely on static exposure estimates and lack spatial granularity. This limits understanding of how pollutant-specific patterns influence psychological states in real-world settings. To address this gap, we integrate real-time environmental and physiological data from 40 participants using the DigitalExposome dataset, applying multivariate and spatial analysis techniques. Our findings confirm that Particulate Matter (PM_2.5) exerts the strongest negative association with mental wellbeing while extending prior work by establishing a preliminary ranking of other pollutants Particulate Matter (PM₁₀), Particulate Matter (PM₁), Carbon Monoxide (CO), Nitrogen Dioxide (NO₂), Ammonia (NH₃). We applied statistical and spatial analysis methods, including heatmaps and Voronoi diagrams, to explore links between pollutants and wellbeing and compare the relative influence of air pollution and noise. This enabled identification of pollutant-specific hotspots and multi-level wellbeing patterns across individual, accumulated, and collective scales. These results demonstrate the value of spatial analysis for environmental health research and support targeted urban interventions, such as green space placement and traffic re-routing, to mitigate mental wellbeing risks. Full article

The rapid deployment of surveillance cameras in urban, industrial, and domestic environments has intensified the need for intelligent systems capable of analyzing video streams beyond the limitations of single-camera setups. Unlike traditional single-camera approaches, multi-camera systems expand spatial coverage, reduce blind spots, and enable consistent tracking of people and objects across non-overlapping views, thereby improving robustness against occlusions and viewpoint changes. This article presents a comprehensive review of multi-camera vision systems published between 2020 and 2025, covering application domains including public security and biometrics, intelligent transportation, smart cities and IoT, healthcare monitoring, precision agriculture, industry and robotics, pan–tilt–zoom (PTZ) camera networks, and emerging areas such as retail and forensic analysis. The review synthesizes predominant technical approaches, including deep-learning-based detection, multi-target multi-camera tracking (MTMCT), re-identification (Re-ID), spatiotemporal fusion, and edge computing architectures. Persistent challenges are identified, particularly in inter-camera data association, scalability, computational efficiency, privacy preservation, and dataset availability. Emerging trends such as distributed edge AI, cooperative camera networks, and active perception are discussed to outline future research directions toward scalable, privacy-aware, and intelligent multi-camera infrastructures. Full article

Background/Objectives: Wound healing is a complex biological process involving inflammatory, proliferative, and remodeling phases. Plant-derived essential oils are increasingly investigated as topical therapeutic agents, although their biological effects are strongly influenced by composition and formulation. The present study evaluated the effects of topical Mentha spicata essential oil on cutaneous wound healing in a rat excisional wound model and explored potential molecular mechanisms using a network-based bioinformatic approach. Methods: Twenty-one male Wistar rats were randomly assigned to three groups and treated twice daily for 14 days with a formulation containing 5% Mentha spicata essential oil diluted in olive oil, olive oil alone, or no treatment. Wound healing was assessed through macroscopic monitoring and histological scoring. The chemical composition of the essential oil was characterized using gas chromatography–mass spectrometry analysis. Predicted molecular targets of the major monoterpenes were analyzed through protein interaction networks and pathway enrichment analysis. Results: Macroscopic wound closure progressed in all groups by day 14. Histological analysis revealed that the olive oil group showed more advanced collagen deposition, re-epithelialization, and granulation tissue maturation, whereas the Mentha spicata group displayed a more pronounced inflammatory and proliferative histological pattern. Network-based analysis highlighted signaling pathways related to receptor-mediated cellular responses as potential molecular mechanisms associated with early inflammatory and proliferative processes. Conclusions: These findings suggest that the biological effects of Mentha spicata essential oil in wound repair may be phase-dependent and influenced by concentration and formulation. The results support further studies aimed at optimizing dose and delivery strategies for essential oil–based wound therapies. Full article

The endothelial protein C receptor (EPCR) is an important component of the protein C (PC) system, recognised for its diverse roles in blood coagulation, inflammation, and stem cell regulation. Wound healing is a complex physiological process that can be divided into four distinct but overlapping phases: haemostasis, inflammation, proliferation and remodelling. Recently, EPCR has emerged as a key regulator in wound repair and regeneration. During haemostasis, EPCR enhances the conversion of PC to its activated form (APC) to optimise local and systemic anticoagulation. In the inflammatory phase, EPCR modulates immune cell activity, inhibits inflammatory factors, and maintains tissue barrier integrity. As the process transitions to the proliferative phase, EPCR promotes endothelial and epithelial cell proliferation, migration, neovascularisation and re-epithelization, and mediates the expression of matrix metalloproteinases to facilitate tissue reconstruction. Finally, during the remodelling phase, EPCR exerts a potential antifibrotic effect by regulating fibroblast activation and collagen deposition via the Transforming growth factor (TGF)-β1/Smad3 pathway, ensuring functional repair. While therapeutic potential has been shown in animal models, translating EPCR-mediated therapies to clinical application faces many challenges, including wound heterogeneity, dosage control, targeted delivery, and potential bleeding risks. Studies have shown that local drug delivery strategies, non-anticoagulant APC variants, and individualised treatment based on EPCR expression will be the key directions for future development. Additionally, EPCR may serve as a potential biomarker for assessing wound severity and guiding personalised interventions. Full article

To address the challenge of detecting weak targets with small radar cross-sections (RCS), this work explores an integrated framework that leverages cross-level multimodal fusion of radar echoes. This method considers the target’s motion properties via Doppler spectrum and phase sequences (direct physical level), and introduces the Gramian Angular Field (GAF) to map the echo amplitude sequence into two-dimensional (2D) structured images, thereby revealing the dynamic evolution characteristics of echo energy (abstract representation level). This approach integrates direct physical attributes and abstract system evolution features within a unified representation. To accommodate the structural differences among modalities, a heterogeneous branch processing network is designed: the Transformer is employed to capture long-range dependencies in one-dimensional (1D) sequences, while ResNet18 is used to extract spatial texture features from two-dimensional images. A self-attention mechanism is further introduced to achieve adaptive fusion of the multimodal data. Experimental results based on the IPIX dataset suggest that this cross-level strategy provides improved detection performance across various scenarios, as observed in complex marine environments. Full article

Maritime ship target tracking plays an important role in applications such as maritime patrol and maritime surveillance. However, complex sea conditions, similar target appearances, and long-distance imaging often lead to target identity confusion and unstable trajectories. To address these issues, in this paper, a ship multi-object tracking algorithm, DeepSORT-OCR, that integrates hull number semantic features is proposed. Based on the YOLO detection framework and the DeepSORT tracking architecture, a CBAM-ResNet network is introduced to enhance the representation of ship appearance features. An Inner-SIoU metric is adopted to improve the geometric matching of slender ship targets, while an LSTM-Adaptive Kalman Filter is employed to model the nonlinear motion patterns of ships and improve trajectory prediction stability. In addition, a Hull Number Feature Extraction module is designed in order to recognize ship hull numbers using OCR and match them with a hull number database. The extracted hull number semantic features are dynamically fused with visual appearance features to strengthen identity constraints during target association. The experimental results show that the proposed method achieves an MOTA of 66.53% on the MOT16 dataset, representing an improvement of 5.13% over DeepSORT. On the self-constructed maritime ship dataset, the method achieves an MOTA of 70.89% and an MOTP of 80.84%. Furthermore, on the hull-number subset, the MOTA further increases to 77.18%, an improvement of 7.31% compared with DeepSORT, while the number of ID switches is significantly reduced. In addition, experiments conducted on pure real data, pure synthetic data, and cross-domain evaluation settings demonstrate the stability and strong generalization capability of the proposed algorithm under different data distributions. The proposed method effectively improves the stability and identity consistency of ship multi-object tracking in complex maritime environments. Full article