Search Results (425)

In contemporary industrial machinery, bearings are a vital component, so the ability to diagnose bearing faults is extremely important. Current methodologies face challenges in feature extraction and perform suboptimally in environments with high noise levels. This paper proposes an enhanced, multimodal, feature-fusion-bearing fault diagnosis model. Integrating a 1DCNN-dual MLP framework with an enhanced two-way cross-attention mechanism enables in-depth feature fusion. Firstly, the raw fault time-series data undergo fast Fourier transform (FFT). Then, the original time-series data are input into a multi-layer perceptron (MLP) and a one-dimensional convolutional neural network (1DCNN) model. The frequency-domain data are then entered into the other multi-layer perceptron (MLP) model to extract deep features in both the time and frequency domains. These features are then fed into a serial bidirectional cross-attention mechanism for feature fusion. At the same time, a GCOS learning rate scheduler has been developed to automatically adjust the learning rate. Following fifteen independent experiments on the Case Western Reserve University bearing dataset, the fusion model achieved an average accuracy rate of 99.83%. Even in a high-noise environment (0 dB), the model achieved an accuracy rate of 90.66%, indicating its ability to perform well under such conditions. Its accuracy remains at 86.73%, even under 0 dB noise and variable operating conditions, fully demonstrating its exceptional robustness. Full article

This paper investigates federated learning (FL) for cross-site scripting (XSS) detection under out-of-distribution (OOD) drift. Real-world XSS traffic involves fragmented attacks, heterogeneous benign inputs, and client imbalance, which erode conventional detectors. To simulate this, we construct two structurally divergent datasets: one with obfuscated, mixed-structure samples and another with syntactically regular examples, inducing structural OOD in both classes. We evaluate GloVe, GraphCodeBERT, and CodeT5 in both centralised and federated settings, tracking embedding drift and client variance. FL consistently improves OOD robustness by averaging decision boundaries from cleaner clients. Under FL scenarios, CodeT5 achieves the best aggregated performance (97.6% accuracy, 3.5% FPR), followed by GraphCodeBERT (96.8%, 4.7%), but is more stable on convergence. GloVe reaches a competitive final accuracy (96.2%) but exhibits a high instability across rounds, with a higher false positive rate (5.5%) and pronounced variance under FedProx. These results highlight the value and limits of structure-aware embeddings and support FL as a practical, privacy-preserving defence within OOD XSS scenarios. Full article

To address the semantic mismatch between limited textual descriptions in image captioning training datasets and the multi-semantic nature of images, as well as the underutilized external commonsense knowledge, this article proposes a novel image captioning model based on multi-step cross-attention cross-modal alignment and external commonsense knowledge enhancement. The model employs a backbone architecture comprising CLIP’s ViT visual encoder, Faster R-CNN, BERT text encoder, and GPT-2 text decoder. It incorporates two core mechanisms: a multi-step cross-attention mechanism that iteratively aligns image and text features across multiple rounds, progressively enhancing inter-modal semantic consistency for more accurate cross-modal representation fusion. Moreover, the model employs Faster R-CNN to extract region-based object features. These features are mapped to corresponding entities within the dataset through entity probability calculation and entity linking. External commonsense knowledge associated with these entities is then retrieved from the ConceptNet knowledge graph, followed by knowledge embedding via TransE and multi-hop reasoning. Finally, the fused multimodal features are fed into the GPT-2 decoder to steer caption generation, enhancing the lexical richness, factual accuracy, and cognitive plausibility of the generated descriptions. In the experiments, the model achieves CIDEr scores of 142.6 on MSCOCO and 78.4 on Flickr30k. Ablations confirm both modules enhance caption quality. Full article

The Tibetan Plateau’s lakes, serving as critical water towers for over two billion people, exhibit divergent responses to climate change that remain poorly quantified. This study develops a deep learning framework integrating Synthetic Aperture Radar (SAR) altimetry from Sentinel-3A with bias-corrected CMIP6 (Coupled Model Intercomparison Project Phase 6) climate projections under Shared Socioeconomic Pathways (SSP) scenarios (SSP2-4.5 and SSP5-8.5, adjusted via quantile mapping) to predict lake-level changes across eight Tibetan Plateau (TP) lakes. Using a Feed-Forward Neural Network (FFNN) optimized via Bayesian optimization using the Optuna framework, we achieve robust water level projections (mean validation R² = 0.861) and attribute drivers through Shapley Additive exPlanations (SHAP) analysis. Results reveal a stark north–south divergence: glacier-fed northern lakes like Migriggyangzham will rise by 13.18 ± 0.56 m under SSP5-8.5 due to meltwater inputs (temperature SHAP value = 0.41), consistent with the early (melt-dominated) phase of the IPCC’s ‘peak water’ framework. In comparison, evaporation-dominated southern lakes such as Langacuo face irreversible desiccation (−4.96 ± 0.68 m by 2100) as evaporative demand surpasses precipitation gains. Transitional western lakes exhibit “peak water” inflection points (e.g., Lumajang Dong’s 2060 maximum) signaling cryospheric buffer loss. These projections, validated through rigorous quantile mapping and rolling-window cross-validation, provide the first process-aware assessment of TP Lake vulnerabilities, informing adaptation strategies under the Sustainable Development Goals (SDGs) for water security (SDG 6) and climate action (SDG 13). The methodological framework establishes a transferable paradigm for monitoring high-altitude freshwater systems globally. Full article

To address the limitation of generalization of federated learning under non-independent and identically distributed (Non-IID) data, we propose FedDFPA, a personalized federated learning framework that integrates dynamic parameter fusion and prototype alignment. We design a class-wise dynamic parameter fusion mechanism that adaptively fuses global and local classifier parameters at the class level. It enables each client to preserve its reliable local knowledge while selectively incorporating beneficial global information for personalized classification. We introduce a prototype alignment mechanism based on both global and historical information. By aligning current local features with global prototypes and historical local prototypes, it improves cross-client semantic consistency and enhances the stability of local features. To evaluate the effectiveness of FedDFPA, we conduct extensive experiments on various Non-IID settings and client participation rates. Compared to the average performance of state-of-the-art algorithms, FedDFPA improves the average test accuracy by 3.59% and 4.71% under practical and pathological heterogeneous settings, respectively. These results confirm the effectiveness of our dual-mechanism design in achieving a better balance between personalization and collaboration in federated learning. Full article

Background/Objectives: Sarcopenic obesity, defined by the coexistence of excessive fat accumulation and progressive muscle loss, is associated with an increased risk of metabolic dysfunction and physical disability. While 5,7-dimethoxyflavone (DMF), a bioactive flavone derived from Kaempferia parviflora, has demonstrated anti-obesity and muscle-preserving properties, its effects on sarcopenic obesity remain unclear. Methods: Four-week-old male C57BL/6J mice were fed a high-fat diet (HFD) for 6 weeks to induce sarcopenic obesity, followed by 8 weeks of continued HFD with the oral administration of DMF. Muscle function was assessed through grip strength and treadmill running tests, while muscle and fat volumes were measured using micro-CT. Mechanistic analyses were performed using gene expression and Western blot analysis. Results: DMF significantly reduced body weight, fat mass, and adipocyte size while enhancing grip strength, endurance, skeletal muscle mass, and the muscle fiber cross-sectional area. In the gastrocnemius muscle, DMF increased the gene expression of peroxisome proliferator-activated receptor gamma coactivator-1α (Ppargc1a) and its isoform Ppargc1a4, thereby promoting mitochondrial biogenesis. It also improved protein turnover by modulating protein synthesis and degradation via the phosphatidylinositol 3-kinase/protein kinase B/mechanistic target of rapamycin signaling pathway. In subcutaneous and brown adipose tissues, DMF increased mitochondrial DNA content and the expression of thermogenic and beige adipocyte-related genes. These findings suggest that DMF alleviates sarcopenic obesity by improving mitochondrial function and regulating energy metabolism in both skeletal muscle and adipose tissues via PGC-1α-mediated pathways. Thus, DMF represents a promising therapeutic candidate for the integrated management of sarcopenic obesity. Full article

The ship waterjet propulsion system is a crucial power unit for high-performance vessels, and the operational state of its core component, the waterjet pump, is directly related to navigation safety and mission reliability. To enhance the intelligence and accuracy of pump fault diagnosis, this paper proposes a novel diagnostic framework that integrates a supervised autoencoder (SAE) with a large language model (LLM). This framework first employs an SAE to perform task-oriented feature learning on raw vibration signals collected from the pump’s guide vane casing. By jointly optimizing reconstruction and classification losses, the SAE extracts deep features that both represent the original signal information and exhibit high discriminability for different fault classes. Subsequently, the extracted feature vectors are converted into text sequences and fed into an LLM. Leveraging the powerful sequential information processing and generalization capabilities of LLM, end-to-end fault classification is achieved through parameter-efficient fine-tuning. This approach aims to avoid the traditional dependence on manually extracted time-domain and frequency-domain features, instead guiding the feature extraction process via supervised learning to make it more task-specific. To validate the effectiveness of the proposed method, we compare it with a baseline approach that uses manually extracted features. In two experimental scenarios, direct diagnosis with full data and transfer diagnosis under limited-data, cross-condition settings, the proposed method significantly outperforms the baseline in diagnostic accuracy. It demonstrates excellent performance in automated feature extraction, diagnostic precision, and small-sample data adaptability, offering new insights for the application of large-model techniques in critical equipment health management. Full article

Callistemon citrinus has shown antioxidant and anti-inflammatory properties in certain tissues. However, its impact on the brain remains unproven. This study investigates the effect of C. citrinus extract and phytosomes on the oxidative status of the brains of rats fed a high-fat–fructose diet (HFD). Fifty-four male Wistar rats were randomly divided into nine groups (n = 6). Groups 1, 2, and 3 received a standard chow diet; Group 2 also received the vehicle, and Group 3 was supplemented with C. citrinus extract (200 mg/kg). Groups 4, 5, 6, 7, 8, and 9 received a high-fat diet (HFD). Additionally, groups 5, 6, 7, 8, and 9 were supplemented with orlistat at 5 mg/kg, C. citrinus extract at 200 mg/kg, and phytosomes loaded with C. citrinus at doses of 50, 100, and 200 mg/kg, respectively. Administration was oral for 16 weeks. Antioxidant enzymes, biomarkers of oxidative stress, and fatty acid content in the brain were determined. A parallel artificial membrane permeability assay (PAMPA) was employed to identify compounds that can cross the intestinal and blood–brain barriers. The HFD group (group 4) increased body weight and adipose tissue, unlike the other groups. The brain fatty acid profile showed slight variations in all of the groups. On the other hand, group 4 showed a decrease in the activities of antioxidant enzymes SOD, CAT, and PON. It reduced GSH level, while increasing GPx activity as well as MDA, 4-HNE, and AOPP levels. C. citrinus extract and phytosomes restore the antioxidant enzyme activities and mitigate oxidative stress in the brain. C. citrinus modulates oxidative stress in brain tissue through 1.8-cineole and α-terpineol, which possess antioxidant and anti-inflammatory properties. Full article

Large Vision Language Models (LVLMs) have shown promise in remote sensing applications, yet struggle with “uncommon” objects that lack sufficient public labeled data. This paper presents Enhanced-RSCLIP, a novel dual-prompt architecture that combines text prompting with exemplar-image processing for cattle herd detection in satellite imagery. Our approach introduces a key innovation where an exemplar-image preprocessing module using crop-based or attention-based algorithms extracts focused object features which are fed as a dual stream to a contrastive learning framework that fuses textual descriptions with visual exemplar embeddings. We evaluated our method on a custom dataset of 260 satellite images across UK and Nigerian regions. Enhanced-RSCLIP with crop-based exemplar processing achieved 72% accuracy in cattle detection and 56.2% overall accuracy on cross-domain transfer tasks, significantly outperforming text-only CLIP (31% overall accuracy). The dual-prompt architecture enables effective few-shot learning and cross-regional transfer from data-rich (UK) to data-sparse (Nigeria) environments, demonstrating a 41% improvement over baseline approaches for uncommon object detection in satellite imagery. Full article

RNA-seq analysis of the highly differentiated human retinal pigment epithelial (RPE) cell-line ARPE-19, cultured on transwells for ≥4 months, yielded 44,909 genes showing 83.35% alignment with the human reference genome. These included mRNA transcripts of RPE-specific genes and those involved in retinopathies. Monolayers were fed photoreceptor outer segments (POS), designed to be synchronously internalised, mimicking homeostatic RPE activity. Cells were subsequently fixed at 4, 6, 24 and 48 h when POS were previously shown to maximally co-localise with Rab5, Rab7, LAMP/lysosomes and LC3b/autophagic compartments. A comprehensive analysis of differentially expressed genes involved in proteolysis revealed a pattern of gene orchestration consistent with POS breakdown in the autophagy-lysosomal pathway. At 4 h, these included elevated upstream signalling events promoting early stages of cargo transport and endosome maturation compared to RPE without POS exposure. This transcriptional landscape altered from 6 h, transitioning to promoting cargo degradation in autolysosomes by 24–48 h. Longitudinal scrutiny of mRNA transcripts revealed nuanced differences even within linked gene networks. POS exposure also initiated transcriptional upregulation in ubiquitin proteasome and chaperone-mediated systems within 4–6 h, providing evidence of cross-talk with other proteolytic processes. These findings show detailed evidence of transcriptome-level responses to cargo trafficking and processing in RPE cells. Full article

This paper presents the design of a wideband circularly polarized crossed-dipole antenna for multi-GNSS applications, covering the frequency range of 1.16–1.61 GHz. The proposed antenna employs orthogonally placed dipole elements fed by a three-branch quadrature hybrid coupler for broadband and wide gain/axial ratio beamwidth. The design is carried out using CST Studio Suite for a single dipole antenna followed by a crossed-dipole antenna, a feed network, and the entire antenna structure. The designed multi-GNSS antenna shows, at 1.16–1.61 GHz, a reflection coefficient of less than −17 dB, a zenith gain of 3.9–5.8 dBic, a horizontal gain of −3.3 to −0.2 dBic, a zenith axial ratio of 0.6–1.0 dB, and horizontal axial ratio of 0.4–5.9 dB. The proposed antenna has a dimension of 0.48 × 0.48 × 0.25 λ at the center frequency of 1.39 GHz. The proposed antenna can also operate as an LHCP antenna for L-band satellite phone communication at 1.525–1.661 GHz. Full article

In this study, the combined influence of vibration direction, feature selection strategy, and the support vector machine (SVM) kernel on the classification accuracy of unbalance faults was investigated. Experiments were carried out on a Jeffcott rotor test rig at a constant speed and under three operating conditions. The overlapping sliding window method was used for raw sample expansion. Features extracted from time domain signals and from the order and power spectra obtained in the frequency domain were ranked using the Kruskal–Wallis algorithm. Based on the feature-ranking results, the three most discriminative features for each domain–axis combination, as well as all nine most discriminative features for each axis in a hybrid manner, were fed into SVM classifiers with different kernels, and their performance was evaluated using ten-fold cross-validation. Classification using vibration signals in the vertical direction had higher accuracy rates than those using signals in the horizontal direction for the feature sets obtained in the same domains. According to the statistical results, feature set selection had a much greater impact on classification accuracy than SVM kernel choice. Power spectrum-based features allowed higher classification accuracies in all SVM algorithms compared to both the time domain features and the order spectrum-based features for detecting unbalance faults. Increasing the number of features or employing hybrid feature selection did not result in a consistent or significant enhancement in overall classification performance. Selecting the right SVM kernel shapes both the model’s flexibility and its fit to the chosen feature space; when this fit is inadequate, classification accuracy may decrease. Consequently, by selecting the appropriate vibration direction, feature set, and SVM kernel, an improvement of up to 67% in unbalance fault classification accuracy was achieved. Full article

Greater amberjack (Seriola dumerili) shows potential for Mediterranean aquaculture due to its swift growth, consumer appeal, and commercial value. However, challenges in juvenile production, such as growth dispersion and unsynchronized development, impede further expansion. This study explores the impact of rearing temperature and live feed types on early white muscle development in greater amberjack larvae. Findings reveal substantial effects of temperature and diet on larval development, highlighting that the combination of 24 °C and a copepod + rotifer co-feeding scheme resulted in the highest axial growth rate, whereas rotifer-fed larvae at 20 °C exhibited a slower pace. Incorporating both histological and gene expression analyses, the study underscores temperature’s significant influence on white muscle development. Among larvae reared at 24 °C, the two live feed types led to phenotypic variations at metamorphosis, with rotifers supporting longer larvae featuring a smaller total cross-sectional area compared to copepods. Gene expression analysis indicates heightened mylpfb and myog expression at 24 °C during early larval stages, suggesting increased hyperplasia and myoblast differentiation. This study highlights the necessity of considering both temperature and feed type in larval rearing practices for optimal muscle development, and further research exploring combined diets during rearing could offer insights to enhance amberjack aquaculture sustainability. Full article

Drought impacts agricultural production and regional sustainable development. Accordingly, timely and accurate drought monitoring is essential for ensuring food security in rain-fed agricultural regions. Alternating drought and flood events frequently occur in the Heilongjiang River Basin, the largest grain-producing area in Far East Asia. However, spatiotemporal variability in drought is not well understood, in part owing to the limitations of the traditional Temperature Vegetation Dryness Index (TVDI). In this study, an Improved Temperature Vegetation Dryness Index (ITVDI) was developed by incorporating Digital Elevation Model data to correct land surface temperatures and introducing a constraint line method to replace the traditional linear regression for fitting dry–wet boundaries. Based on MODIS (Moderate-resolution Imaging Spectroradiometer) normalized vegetation index and land surface temperature products, the Heilongjiang River Basin, a cross-border basin between China, Mongolia, and Russia, exhibited pronounced spatiotemporal variability in drought conditions of the growing season from 2001 to 2023. Drought severity demonstrated clear geographical zonation, with a higher intensity in the western region and lower intensity in the eastern region. The Mongolian Plateau and grasslands were identified as drought hotspots. The Far East Asia forest belt was relatively humid, with an overall lower drought risk. The central region exhibited variation in drought characteristics. From the perspective of cross-national differences, the drought severity distribution in Northeast China and Inner Mongolia exhibits marked spatial heterogeneity. In Mongolia, regional drought levels exhibited a notable trend toward homogenization, with a higher proportion of extreme drought than in other areas. The overall drought risk in the Russian part of the basin was relatively low. A trend analysis indicated a general pattern of drought alleviation in western regions and intensification in eastern areas. Most regions showed relatively stable patterns, with few areas exhibiting significant changes, mainly surrounding cities such as Qiqihar, Daqing, Harbin, Changchun, and Amur Oblast. Regions with aggravation accounted for 52.29% of the total study area, while regions showing slight alleviation account for 35.58%. This study provides a scientific basis and data infrastructure for drought monitoring in transboundary watersheds and for ensuring agricultural production security. Full article

This paper proposes and demonstrates a domain-adapted ensemble machine learning approach for enhanced prediction of surface roughness (Ra) during the machining of polymeric materials. The proposed model methodology employs a two-stage pipelined architecture, where classified data are fed into the model for regressive analysis. First, a classifier (Logistic Regression or XGBoost, selected based on performance) categorizes machining data into distinct regimes based on cutting Speed (Vc), feed rate (f), and depth of cut (ap) as inputs. This classification leverages output discretization to mitigate data imbalance and capture regime-specific patterns. Second, a regressor (Support Vector Regressor or XGBoost, selected based on performance) predicts Ra within each regime, utilizing the classifier’s output as an additional feature. This structured hybrid approach enables more robust prediction in small, noisy datasets characteristic of machining studies. To validate the methodology, experiments were conducted on Polyoxymethylene (POM), Polytetrafluoroethylene (PTFE), Polyether ether ketone (PEEK), and PEEK/MWCNT composite, using a L₂₇ Design of Experiments (DoEs) matrix. Model performance was optimized using k-fold cross-validation and hyperparameter tuning via grid search, with R-squared and RMSE as evaluation metrics. The resulting meta-model demonstrated high accuracy (R² > 90% for XGBoost regressor across all materials), significantly improving Ra prediction compared to single-model approaches. This enhanced predictive capability offers potential for optimizing machining processes and reducing material waste in polymer manufacturing. Full article