Search Results (16,470)

High-resolution remote sensing imagery encompasses complex land cover types and rich textural details, whilst temporal variations often manifest as subtle feature differences and unstable structural patterns. This renders traditional change detection methods ineffective at accurately characterizing genuine alterations, frequently leading to underdetection, false positives, and ambiguous boundaries. To address these challenges, this paper proposes a Transformer-Based Multi-level Adaptive Fusion Network. It is built upon the DeepLabV3+ encoder–decoder framework, in which a shared-weight ResNet-101 is adopted as the backbone for dual-temporal feature extraction, with the final residual block of layer 4 cropped to extract deeper semantic features at a higher spatial resolution. The Adaptive Window–Attention Feature Fusion Module (AWAFM) adaptively models local and global differences across temporal phases, enhancing sensitivity to genuine changes. The Dual Strip Pool Fusion Module (DSPFM) enhances sensitivity to directional structural variations through horizontal and vertical strip pooling. The Progressive Multi-Scale Feature Fusion Module (PMFFM) progressively aggregates deep and shallow features via semantic residual transmission. To further suppress misleading suppression caused by complex textures, the Transformer-Enhanced Reverse Attention Fusion Module (TRAFM) explicitly models long-range dependencies, effectively mitigating false change responses. On the LEVIR-CD dataset, it achieves state-of-the-art performance, with a PA and an IoU of 92.36% and 90.13%, respectively. On the SYSU-CD dataset, PA and IoU reach 88.96% and 86.15%, demonstrating TMAFNet’s stability and superiority in scenarios involving complex ground surface disturbances, weak textural variations, and large-scale structural changes. Full article

Diabetic retinopathy (DR) remains a major cause of vision loss in patients with diabetes, and earlier recognition of retinal vascular abnormalities may improve risk stratification and clinical follow-up. Optical coherence tomography angiography (OCTA) provides a noninvasive way to visualize the retinal microvasculature and may detect DR-related changes before they are evident on routine clinical assessment. In this work, we investigated whether dividing OCTA images into anatomically defined retinal regions could improve DR classification and clarify which regions carry the greatest discriminative information. The study included 188 OCTA images: 67 from normal eyes, 57 from eyes with mild DR, and 64 from eyes with moderate DR. Each image was divided into seven concentric regions centered on the fovea, and vessel-density features were extracted from each region. Ten machine learning classifiers were trained and compared at the regional level. For each region, the best-performing classifier was retained, and the final prediction was obtained with a majority-voting ensemble. To examine model behavior, Local Interpretable Model-Agnostic Explanations (LIME) were applied. Performance was also compared with that of a transfer-learning MobileNet model trained on whole OCTA images. On the held-out patient-level test set, the ensemble model achieved 97% accuracy, 98% precision, 97% recall, and a 97% F1-score for three-class classification. These results were higher than those obtained with the tested whole-image transfer-learning baselines. The interpretability analysis consistently identified the parafoveal regions as the most informative for classification. Among the seven regions, Region 3 showed the highest overall contribution, followed by Regions 2 and 5, whereas Region 5 became more influential in moderate DR. These results suggest that regional analysis of OCTA-derived vessel density can improve both classification performance and interpretability in DR assessment. The findings also indicate that parafoveal vascular alterations carry substantial discriminative value in distinguishing normal, mild DR, and moderate DR cases. Validation in larger, independent cohorts from multiple centers will be necessary to confirm the generalizability of these findings. Full article

The high morbidity and mortality of cancer pose a severe challenge to human health. Traditional diagnostic and therapeutic strategies still exhibit obvious limitations in early diagnostic sensitivity, therapeutic precision, and real-time monitoring of treatment efficacy. The development of nanotechnology has provided novel solutions for precision cancer theranostics. Among nanomaterials, gold nanoparticles (AuNPs) have become a research hotspot in tumor nanomedicine due to their tunable size and morphology, excellent localized surface plasmon resonance (LSPR) effect, and favorable biocompatibility. However, despite encouraging preclinical outcomes, several challenges hinder their clinical translation, including an incomplete understanding of long-term toxicity, complex in vivo biological interactions, the lack of standardized evaluation protocols, and regulatory uncertainties and manufacturing reproducibility issues. This paper systematically reviews the physicochemical and biological mechanisms of AuNPs in cancer theranostics, and summarizes the latest research advances of AuNPs in cancer detection and diagnosis (including biomarker detection and multimodal imaging) as well as in therapeutic fields, covering photothermal therapy (PTT), photodynamic therapy (PDT), radiosensitization, targeted drug and nucleic acid delivery, and immunotherapy-assisted strategies. Furthermore, we discuss the development of intelligent and stimuli-responsive theranostic nanoplatforms based on AuNPs, and outline their future prospects in precision medicine and personalized cancer therapy, with particular emphasis on the requirements for clinical translation, including safety evaluation, large-scale production, and regulatory approval pathways. Full article

Ferroptosis is an iron-dependent form of regulated cell death driven by lipid peroxidation and oxidative stress, increasingly implicated in cancer biology. However, its molecular regulation across breast cancer subtypes and its potential systemic manifestations remain incompletely understood. The aim of this study was to identify ferroptosis-associated molecular alterations that are largely shared across subtypes and to evaluate their systemic reflection following localized tissue injury. Tumor and matched normal breast tissues representing major molecular subtypes were analyzed. Global mRNA and miRNA expression profiling was performed using microarrays, followed by validation of selected genes using quantitative reverse transcription polymerase chain reaction (qRT-PCR) and enzyme-linked immunosorbent assay (ELISA). Functional enrichment and protein–protein interaction analyses were conducted to characterize associated pathways. In addition, systemic responses were assessed in patients undergoing fibroadenoma cryoablation through longitudinal blood sampling. Six ferroptosis-related genes (SLC7A11, GPX4, FTH1, NQO1, NFE2L2, SQSTM1) demonstrated consistent upregulation across all breast cancer subtypes, with higher expression observed in more aggressive tumors. These genes are functionally linked to antioxidant defense, iron metabolism, and oxidative stress regulation, and their coordinated expression pattern is consistent with activation of NRF2-dependent cytoprotective pathways. Downregulation of selected miRNAs may contribute to this expression profile but likely represents a secondary regulatory mechanism. Survival analysis revealed heterogeneous and subtype-dependent associations, with limited and gene-specific prognostic relevance. Cryoablation induced transient increases in circulating levels of the analyzed proteins, reflecting systemic responses to localized tissue injury. In conclusion, breast cancer is characterized by a largely shared ferroptosis-associated molecular signature across subtypes; however, its clinical impact appears to be variable and context-dependent. Systemic detection of related molecular signals suggests potential utility as indicators of tissue stress responses, although their role as specific biomarkers of ferroptosis requires further validation. Full article

Ameloblastoma (AM) is a common locally invasive benign odontogenic tumor in Asian populations. Although fibroblast growth factor receptor 2 (FGFR2) mutations have been reported in AM, FGFR2 amplification, the predominant form of FGFR2 aberration in human cancers, remains unexplored. This study aimed to evaluate FGFR2 protein expression, FGFR2 gene copy number variations, and their associations with demographic and clinico-radio-pathological parameters in mandibular AM. Eighty-seven cases of mandibular AM and 10 dental follicle (DF) specimens were included. FGFR2 protein expression was assessed by immunohistochemistry, and gene copy number variations were analyzed using the quantitative real-time polymerase chain reaction (qPCR) technique. Clinical data, including age, gender, tumor size, radiographic features, histological subtype, and recurrence history, were examined for associations with FGFR2 alterations. FGFR2 protein overexpression was observed in 95.4% of AM cases and was not significantly associated with demographic or clinico-radio-pathological variables. FGFR2 gene amplification was detected in 52.5% of cases, while 8.2% showed gene deletion. Notably, 50.8% of cases exhibited concurrent FGFR2 amplification and overexpression, and all cases with FGFR2 gene deletion also demonstrated FGFR2 overexpression. These findings suggest that FGFR2 gene amplification and deletion may contribute to FGFR2 overexpression and play a significant role in the molecular pathogenesis of mandibular AM. Full article

Accurate vegetable disease detection in complex cultivation environments remains challenging because early lesions are often small, low-contrast, and easily confounded by cluttered backgrounds. To address this issue, we propose VDD-Net, a feature-enhanced detection network based on YOLOv10 for robust vegetable disease detection in protected agriculture. The proposed framework integrates three modules: a receptive field enhancement (RFE) module to improve local perception of small lesions, an adaptive channel fusion (ACF) module to strengthen multi-scale feature aggregation and suppress background interference, and a global context attention (GCA) module to capture long-range dependencies and improve contextual discrimination. Experiments on a custom vegetable disease dataset showed that VDD-Net achieved an mAP@0.5 of 95.2% with only 7.78 M parameters. To further evaluate robustness, zero-shot cross-domain testing was conducted on the PlantDoc dataset, where VDD-Net achieved an mAP@0.5 of 76.5%, outperforming the baseline and showing improved generalization to natural scenes. In addition, after TensorRT optimization and FP16 quantization, the model maintained real-time inference on edge platforms, reaching 89.3 FPS on Jetson AGX Orin and 24.2 FPS on Jetson Nano. These results indicate that VDD-Net provides a practical balance among detection accuracy, cross-domain robustness, and deployment efficiency for intelligent disease monitoring in modern agriculture. Full article

The accurate and rapid prediction of hazard gas dispersion fields in urban environments is essential for guiding emergency sensor deployment and enabling real-time risk assessment. However, the computational cost associated with Computational Fluid Dynamics (CFD) simulations hinders their use as real-time forward models, while simplified Gaussian plume models lack the fidelity to resolve building obstruction effects. This study proposes a morphology-guided conditional Generative Adversarial Network (cGAN) framework designed to achieve real-time gas dispersion field modeling in urban environments with complex building configurations. The urban area is discretized into 50 × 50 m grid cells, each characterized by six morphological parameters describing building geometry. K-means clustering categorizes these cells into distinct morphological types. High-fidelity dispersion datasets are then generated for each type using Lattice Boltzmann Method (LBM) simulations. Each sample encodes building geometry, release location, wind speed, and time as multi-channel input images, with the corresponding gas dispersion concentration field is recorded as the output. Two cGAN architectures, Image-to-Image Translation (Pix2Pix) and its high-resolution variant (Pix2PixHD), are employed to learn the mapping from input features to dispersion fields. Model performance is evaluated using four complementary metrics: Fraction within a Factor of Two (FAC2) for prediction accuracy, Normalized Root Mean Square Error (NRMSE) for precision, Fractional Bias (FB) for systematic error, and Structural Similarity Index (SSIM) for spatial pattern fidelity. A case study is conducted across a 1176 km² urban district in China. The results demonstrate that under varying wind speeds (0.5–1.5 m/s) and temporal scales (5–60 s), and across five morphological categories, the Pix2PixHD-based model achieves 92.5% prediction accuracy and reproduces 97.6% of the spatial patterns. The proposed framework accelerates computation by approximately 18,000 times compared to traditional CFD, reducing inference time to under 0.1 s per scenario. This sub-second capability enables real-time concentration field estimation for emergency management, and provides a physically informed, computationally feasible forward model that can potentially support sensor-based gas source localization and detection network planning in complex urban environments. Full article

Conventional gas direction estimation methods that rely on concentration gradients or time-of-arrival differences typically require multiple spatially dispersed sensors, leading to increased system bulkiness and complexity. Furthermore, previous CMOS-based approaches that relied on gas diffusion struggled to achieve stable direction estimation in high-speed airflow environments. To address these challenges, we propose a streamlined method integrating a pillar onto a single CMOS gas-sensor array, eliminating additional MEMS fabrication. This approach exploits a fluid dynamic phenomenon where the pillar creates a distinct flow “shadow” pattern (a region of localized gas dilution) on the sensor surface. Experimental verification using ammonia gas confirmed that this “shadow” is clearly observable as a localized reduction in sensor output under high-speed turbulent flow. Crucially, the spatial position of this pattern correlates strongly with the direction of gas inflow. This study demonstrates the feasibility of gas direction estimation using a single chip, paving the way for high-precision detection in challenging, rapid-airflow environments. Full article

Motion blur degrades single-frame imaging when relative motion occurs during sensor exposure; yet, quantitative validation is difficult because ground-truth motion parameters are rarely available in real images. This paper presents an interpretable, measure-first framework for detecting, localizing, and quantifying motion blur in single-frame grayscale images under a validated operating condition of one-dimensional horizontal uniform motion. The method analyzes each image row as a one-dimensional spatial signal, where Movement Artifact denotes the scanline-level imprint of motion blur retained in the legacy algorithm names MAPE and MAQ. The pipeline combines three stages: Movement Artifact Position Estimation (MAPE) using scanline self-similarity, Reference Origin Point Estimation (ROPE) using robust structural trends, and Movement Artifact Quantification (MAQ), which summarizes blur magnitude as an average horizontal spatial displacement after adaptive filtering. The pipeline is evaluated on a controlled empirical dataset of 110 images of a high-contrast marker acquired at known tangential velocities from 0.0 to 1.0 m/s in 0.1 m/s increments (10 images per level). MAPE achieves 70–90% detection rates across velocities, and ROPE localizes reference origins with 97–99% detection. An empirical polynomial mapping from MAQ to velocity attains $R^2=0.9900$ with RMSE 0.0229 m/s and MAE 0.0221 m/s over 0.0–0.7 m/s, enabling calibrated velocity estimates from blur measurements within the validated regime. An extended additive-noise robustness analysis further shows that severe perturbation can preserve candidate self-similarity responses while progressively destabilizing reference-origin localization and MAQ pairing, thereby clarifying the empirical boundary of the current controlled single-marker regime. The approach is not claimed to generalize to uncontrolled scenes, non-uniform blur, or multi-dimensional and non-rigid motion. Full article

Labor shortages and the need for increased productivity have accelerated the development of robotic harvesting systems for greenhouse crops; however, reliable operation under fruit occlusion and clustered arrangements remains a major challenge, particularly due to the limited integration between perception and motion planning modules. The paper presents the design and experimental validation of an autonomous robotic system for greenhouse tomato harvesting. The proposed platform integrates a rail-guided mobile base, a six-degrees-of-freedom robotic manipulator, and an adaptive end effector with a hybrid vision framework that combines convolutional neural networks and watershed-based segmentation to enable robust fruit detection and localization under occluded conditions. The proposed approach enables improved separation of overlapping fruits and provides accurate spatial localization through stereo vision combined with IMU-assisted camera-to-robot coordinate transformation. An occlusion-aware trajectory planning strategy was developed to generate collision-free manipulation paths in the presence of leaves and stems, enhancing harvesting safety and reliability. The system was trained and evaluated using a dataset of real greenhouse images supplemented with synthetic data augmentation. Experimental trials conducted under practical greenhouse conditions demonstrated a fruit detection precision of 96.9%, recall of 93.5%, and mean Intersection-over-Union of 79.2%. The robotic platform achieved an overall harvesting success rate of 78.5%, reaching 85% for unobstructed fruits, with an average cycle time of 15 s per fruit in direct harvesting scenarios. The rail-guided mobility significantly improved positioning stability and repeatability during manipulation compared with fully mobile platforms. The results confirm that integrating hybrid perception with occlusion-aware motion planning can substantially improve the functionality of robotic harvesting systems in protected cultivation environments. The proposed solution contributes to the advancement of automation technologies for greenhouse vegetable production and supports the transition toward more sustainable and labor-efficient agricultural practices. Full article

The adoption of instrumented wheelsets on diagnostic trains offers the possibility of continuous monitoring of wheel–rail contact forces. The collection of large datasets can be exploited for diagnostic purposes, aiming to localize specific track defects, allowing significant improvements in terms of safety and maintenance costs. Machine learning (ML) techniques can be used to automate anomaly detection. In this work, the authors compare the application of various ML algorithms based on the identification of different frequency or time-based features of analyzed signals. To perform the activity, a significant number and variety of local defects have been included in the recorded data. From a practical point of view, the insertion of real known defects into an existing line is extremely time-consuming, expensive, and not immune to safety issues. On the other hand, the design of anomaly detection algorithms involves the usage of relatively extended datasets with different faulty conditions. The authors propose deliberately adding real contact force profiles of healthy lines to a mix of synthetic signals, which substantially reproduce the behavior and the variability of foreseen faulty conditions. The results of this work, although preliminary and still to be completed, offer a contribution to the scientific community both in terms of obtained results and adopted methodologies. Full article

Deep learning is increasingly integrated into oral rehabilitation workflows, particularly in implant planning, prosthodontic design automation, and peri-implant diagnosis. However, reported performance is heterogeneous and difficult to compare across tasks, modalities, and validation designs. The goal of this study was to critically analyze deep learning architecture families applied to oral rehabilitation and to provide task-driven selection guidance supported by an evidence table reporting dataset characteristics, validation strategy, and performance metrics. A focused narrative review was conducted using transparent, database-specific search criteria (final n = 10 included studies), emphasizing implant planning (cone–beam computed tomography [CBCT]-based segmentation), prosthodontic design (intraoral scan [IOS]/mesh inputs), and peri-implant diagnosis (periapical/panoramic radiographs). Evidence certainty for each clinical task was assessed using GRADE-informed ratings (High/Moderate/Low/Very Low). Extracted variables included clinical task, imaging modality, dataset size, architecture, validation strategy (internal vs. internal + external), split level, ground truth protocol, and performance metrics. A structured computational and hardware feasibility analysis was conducted for each architecture family to support real-world deployment planning. Encoder–decoder networks (U-Net/nnU-Net) dominate CBCT segmentation for implant planning, while detection architectures (Faster R-CNN, YOLO) support implant localization and peri-implant assessment on radiographs. Generative models (3D GANs, transformer-based point-to-mesh networks) enable crown design from three-dimensional scans. Hybrid CNN–Transformer architectures show promise for multimodal CBCT–IOS fusion, though direct evidence from the included studies remains limited to a single study. External validation remains uncommon yet essential given the risk of domain shift. In conclusion, architecture selection should be anchored to task geometry (2D vs. 3D), artifact burden, and required clinical output type. Reporting standards should prioritize dataset transparency, validation rigor, multi-center external testing, and uncertainty-aware outputs. Full article

Federated Learning (FL) has been widely adopted in privacy-sensitive and distributed environments. However, training stability becomes significantly challenged when differential privacy (DP) noise and Byzantine client behaviors coexist, as these heterogeneous perturbations jointly introduce time-varying distortions to model updates. Existing approaches typically address privacy and robustness in isolation. Under DP constraints, noise injection increases gradient variance and obscures the distinction between benign and adversarial updates, causing many robust aggregation methods to misclassify normal clients or fail to detect malicious ones. As a result, their effectiveness degrades substantially in practical IoT environments where noise and attacks interact. In this work, we propose a dual-factor adaptive and robust aggregation framework (DARA) to improve the stability of FL under such combined disturbances. DARA adjusts the differential privacy noise scale by jointly considering local update magnitudes and training-round dynamics, aiming to mitigate noise-induced bias under a fixed privacy budget. Meanwhile, a direction-aware weighted aggregation scheme assigns continuous trust weights based on cosine similarity between updates, thereby suppressing the influence of potentially anomalous or adversarial clients. We conduct extensive experiments on multiple benchmark datasets to evaluate DARA under differential privacy constraints and Byzantine attack scenarios. The results indicate that DARA achieves favorable robustness and convergence behavior compared with representative aggregation baselines, while maintaining competitive model accuracy. Full article

Elaeagnus mollis oilseed (EMO) meal is a protein-rich by-product that may serve as a novel source of food-derived α-glucosidase inhibitory peptides. This study aimed to obtain EMO peptide fractions with enhanced α-glucosidase inhibition and to clarify the activity, stability and mechanism of the most active fraction. Fourteen proteases were compared, and 3.350 acidic protease was selected to establish an optimized hydrolysis process. The resulting EMO hydrolysate showed an IC₅₀ of 9.11 mg/mL against α-glucosidase and no detectable cytotoxicity towards HEK-293T cells at 0.1–12.0 mg/mL. Ultrafiltration yielded four fractions, among which the 3–10 kDa fraction exhibited the highest inhibition and maintained substantial activity under acidic pH (2–6), −20–50 °C, NaCl ≤ 5% and simulated gastrointestinal digestion. Kinetic analysis indicated mixed-type inhibition, while fluorescence, circular dichroism and molecular docking suggested that peptides in this fraction bind near the catalytic site of α-glucosidase and induce local conformational changes. These findings support EMO-derived 3–10 kDa peptides as stable, non-cytotoxic α-glucosidase inhibitors with potential as functional ingredients for dietary management of type 2 diabetes. Full article

Seasonal human coronaviruses (HCoVs) continue to undergo adaptive evolution under structural and immune-mediated constraints. We investigated the molecular epidemiology and spike (S) protein structural variation of circulating coronaviruses in Riyadh, Saudi Arabia, during the 2025–2026 winter season, with particular emphasis on genotype persistence and glycosylation architecture in HCoV-OC43. Among 293 nasopharyngeal aspirates (NPAs) collected from hospitalized patients with acute respiratory illness, HCoV-OC43 was detected in 26 cases (8.87%), whereas other seasonal coronaviruses were not identified. Partial sequencing of the S gene revealed 97.84–98.23% nucleotide identity relative to the prototype strain VR-759, with amino acid substitutions distributed at discrete positions rather than within extended variable domains, indicating structural conservation. Phylogenetic reconstruction demonstrated that all Riyadh isolates clustered within genotype C, together with previously circulating local strains, supporting sustained endemic persistence and in situ evolution. In silico analysis of the S protein glycosylation landscape identified four invariant N-linked glycosylation motifs (N-X-S/T) at residues 46, 121, 134, and 190, reflecting strong structural constraints on glycan-dependent folding and antigenic configuration. A genotype-associated K68N substitution generated an additional N-glycosylation motif (68NGTD) in multiple Riyadh isolates, potentially modifying local glycan shielding without disrupting the overall glycosylation framework. The preservation of core glycosylation sites alongside selective motif acquisition suggests evolutionary fine-tuning of S surface topology rather than large-scale structural remodeling. Collectively, these findings indicate that genotype C persistence in Riyadh is accompanied by conserved S architecture and subtle glycosylation adjustments that may modulate immune recognition while maintaining structural integrity. Continued high-resolution molecular surveillance will be critical for defining the functional consequences of S microevolution in endemic HCoVs. Full article