Search Results (1,818)

Unsupervised anomalous sound detection aims to learn acoustic features solely from the operational sounds of normal equipment and identify potential anomalies based on these features. Recent self-supervised classification frameworks based on machine ID metadata have achieved promising results, but they still face two challenges in industrial acoustic scenarios: Log-Mel spectrograms tend to weaken high-frequency details, leading to insufficient spectral characterization, and when normal sounds from different machine IDs are highly similar, classification constraints alone struggle to form clear intra-class structures and inter-class boundaries, resulting in false positives. To address these issues, this paper proposes FGASpecNet, an anomaly detection model integrating spectral enhancement and frequency-gated attention. For feature modeling, a spectral enhancement branch is designed to explicitly supplement spectral details, while a frequency-gated attention mechanism highlights key frequency bands and temporal intervals conditioned on temporal context. Regarding loss design, a joint training strategy combining classification loss and metric learning loss is adopted. Multi-center prototypes enhance intra-class compactness and inter-class separability, improving detection performance in scenarios with similar machine IDs. Experimental results on the DCASE 2020 Challenge Task 2 for anomalous sound detection demonstrate that FGASpecNet achieves 95.04% average AUC and 89.68% pAUC, validating the effectiveness of the proposed approach. Full article

Background/Objectives: The KRAS rs61764370 T>G single-nucleotide polymorphism (SNP), located in a let-7 microRNA binding site within the 3′ untranslated region (3′UTR) of the KRAS gene, may modulate tumor aggressiveness by altering post-transcriptional gene regulation. This study evaluated its association with adverse histopathological features in colorectal cancer (CRC). Methods: A preliminary study on 83 CRC patients carrying either the TT (wild-type, n = 64) or TG (heterozygous, n = 19) genotype was analyzed. Clinicopathological variables included patient sex, tumor location, American Joint Committee on Cancer (AJCC) staging system, histological grade, perineural invasion (PNI), and lymphovascular invasion (LVI). A composite “tumor aggressiveness” score was defined based on the presence of Grade 3 differentiation, LVI, and/or PNI. Group comparisons were performed using the Chi-square test or Fisher’s exact test, as appropriate. Results: No statistically significant differences were observed in sex (p = 0.689), tumor location (p = 0.781), or stage at diagnosis (p = 0.812). Poorly differentiated tumors (Grade 3) were present in 20.3% of TT patients and absent in TG carriers (p = 0.06), while low-grade tumors (Grade 1) were more prevalent among TG patients (47.4%) compared to TT (29.7%). The composite high-aggressiveness score was lower in TG (36.8%) than in TT (48.4%), while co-occurrence of PNI and LVI was similar in both groups (~26%). Conclusions: Although no significant associations were identified, TG carriers showed a tendency toward lower-grade, less aggressive tumors. Given the limited sample size, these findings should be interpreted with caution, necessitating larger cohorts in order to validate results. Full article

Fatigue crack propagation in friction stir welded joints significantly affects aircraft structural integrity. This study investigates the influence of welding speed, rotational speed, specimen thickness, loading frequency, and stress ratio on the fatigue crack growth rate. Four classical machine learning models with different structures—Deep Back-Propagation Network, Random Forest, Support Vector Regression, and K-Nearest Neighbors—were employed to predict fatigue crack growth behavior. The results show that all models achieve strong predictive performance. For FSWed joints, Deep BP and KNN exhibit comparable performance (R² > 0.98) on the training data, indicating similar learning capabilities with sufficient data coverage. Notably, KNN achieves the fastest training time (<0.3 s), while all models require less than 5 s of computation time. These results demonstrate that machine learning-based models provide an efficient and reliable alternative for rapid fatigue crack growth evaluation, supporting damage-tolerant design and structural integrity assessment in aircraft engineering. Full article

As a critical component of demolition robots, the rotary joint supports the entire manipulator arm and operates under severe loading conditions, rendering it highly susceptible to fatigue failure. To address this challenge, topology optimization is integrated into the structural design to simultaneously enhance fatigue life and achieve lightweighting. In this work, multiple working conditions of the demolition robot are considered and analyzed to identify the extreme operating condition. By extracting the resultant stress on the rotary joint from the assembled structure under the extreme condition, an equivalent model of the independent rotary joint is established. Given that topology optimization based on the original structure could not yield a usable structure, two topology optimization strategies based on resetting the design space are proposed, including topology optimization based on the partially filled design space and topology optimization within the fully filled design space. By performing topology optimization under different schemes, the optimized rotary joint models are reconstructed through geometric fusion. Numerical results demonstrate that the optimized rotary joints exhibit significant improvements in fatigue performance, with fatigue life doubled compared to the original design. Concurrently, the structural mass is effectively reduced. This proposed method achieves the dual objectives of fatigue life enhancement and lightweight design. Furthermore, the results reveal that resetting the design space when topology optimization fails to obtain a usable structure yields superior topology optimization outcomes, providing a valuable new insight for future structural optimization design processes in similar engineering scenarios. Full article

Background: Spondyloarthritis (SpA) typically develops before 40 years of age, but increasing life expectancy has led to a growing number of cases in older adults. It is well known that age at onset may influence disease presentation, comorbidities, and patient outcomes. Objectives: To assess whether age at onset influences SpA clinical presentation. Methods: We analyzed clinical, demographic, clinimetric, and imaging data in 272 SpA patients, grouped by onset age: early (≤40, n = 119), intermediate (41–59, n = 127), and late (≥60, n = 26). All patients had a minimum follow-up duration of 12 months. Their epidemiologic, clinic, and clinimetric data were collected, as well as patient-reported outcome measures (PROs) [Patient Global Assessment (PGA), Health Assessment Questionnaire (HAQ), FACIT-Fatigue (FACIT-F), SHORT-FORM 36 (SF-36), Hospital Anxiety and Depression Scale (HADS), Work Productivity and Activity Impairment Questionnaire (WPAI), CSI (Central Sensitization Inventory), and Psoriatic Arthritis Impact of Disease (PsAID) questionnaire]. In univariate analyses, differences in categorical variables across onset groups were assessed using Fisher’s exact test; for continuous variables, between-group comparisons were performed using the Mann–Whitney U test (two-tailed) or the Kruskal–Wallis test, as appropriate, with Bonferroni correction for post hoc analyses. Multivariable regression models were subsequently fitted, adjusting for sex, diagnosis, and disease duration. For binary outcomes, multivariable logistic regression models were used, while multivariable linear regression models (ANCOVA) were applied for continuous outcomes. The overall association between onset group and each outcome was formally tested using likelihood ratio tests, comparing models including the onset variable with nested models excluding it. A p-value < 0.05 was considered statistically significant. Results: Patients’ mean age was 60.0 ± 13.7 years; 55.9% of them were males; and there were 188 cases (69.1%) of psoriatic arthritis (PsA) and 84 cases (30.9%) of ankylosing spondylitis (AS). In early-onset patients, inflammatory back pain (IBP) was more frequent, whereas late-onset patients more often presented with joint swelling. A family history of SpA and psoriasis was less common in late-onset forms. Comorbidities, including osteoporosis, osteoarthritis, hypertension, hyperuricemia, and diabetes, were more prevalent in older-onset patients, resulting in a higher overall comorbidity burden in Groups 2 and 3. Patient-reported outcomes were largely similar across age groups, although work activity limitation was more pronounced in younger patients. Conclusions: Age at onset seems to influence SpA phenotypes: early-onset could favor axial involvement, while late-onset may associate with peripheral arthritis. Late-onset forms are associated with a more severe comorbidity burden, in particular for cardiovascular risk factors. Lung involvement proved to be more prevalent with respect to the general population, so it should be checked in the routinary assessment of SpA patients. These findings suggest that rheumatologists could tailor their routine assessments based on patients’ age at disease onset. Interestingly, work productivity seems more impacted in early-onset patients. All these points highlight the importance of age at disease onset in SpA, guiding toward personalized medicine in terms of follow-up, therapy, and more holistic patient management. Full article

Constrained by observation conditions and high inter-class similarity, effective feature extraction and classification of synthetic aperture radar (SAR) targets in few-shot scenarios remains a persistent challenge. To address this issue, this article proposes a few-shot fine-grained SAR target classification method based on multi-scale embedding network and meta-contrastive learning (MSMC). Specifically, the MSMC integrates two complementary training pipelines; the first employs metric-based meta-learning to facilitate few-shot classification, while the second adopts an auxiliary training strategy to enhance feature diversity through contrastive learning. Furthermore, a shared multi-scale embedding network (MSEN) is designed to extract discriminative multi-scale features via adaptive candidate region generation and joint multi-scale embedding. The experimental results on the MSTAR dataset demonstrate that the proposed method achieves superior few-shot fine-grained classification performance compared to existing methods. Full article

Background: The clinical meaning of hip joint synovial cavity thickness (HJSCT) on ultrasound (US) in juvenile idiopathic arthritis (JIA) without effusion is uncertain. Methods: In this cross-sectional study, we analyzed 369 children (187 JIA; 182 controls) undergoing hip US at a referral center in Kraków, Poland. JIA examinations were performed upon initial referral, early in the care pathway. We excluded patients with hip effusion and pre-existing inflammatory, traumatic or degenerative hip pathology. HJSCT was defined as the distance from the outer capsule margin to the femoral neck cortex. We used a Toshiba Aplio 400 system with a 12 MHz probe to measure and derive mean bilateral HJSCT. Bilateral concordance was assessed. Iterative multivariable linear regression modeling was used to compare groups, adjusting for non-linear age effects (natural splines) and WHO height-for-age z-scores (HAZ). Results: Left–right HJSCT agreement was high (ICC 0.947; mean difference 0.03 mm; 95% limits of agreement −0.64–0.70). In unadjusted analysis, mean (SD) HJSCT was similar in JIA versus controls: 5.83 (1.09) vs. 5.95 (0.99) mm, respectively (p = 0.25). In the final model (adj. R2 0.656), HJSCT was strongly associated with age (non-linear, p < 0.001) but not significantly associated with HAZ (β = 0.04; p = 0.11) or JIA status (β = 0.07; p = 0.30). Predicted HJSCT showed a steep increment in childhood and plateau in adolescence. Conclusions: In children without hip effusion, HJSCT mainly reflects physiological growth and does not differ significantly between early JIA patients and healthy controls. These findings suggest that capsular thickening is not a reliable standalone marker for early disease in the absence of effusion. Full article

Joint roughness coefficient (JRC) and inclination exert a decisive influence on the stability and safety of rock mass engineering. Simulations of uniaxial compression were conducted on saw-tooth-shaped joint specimens using a calibrated particle flow (PFC2D) model. The specimens contained five JRC values (0, 5, 10, 15, 20) and five joint inclinations (0°, 30°, 45°, 60°, 90°). The results indicate that at joint inclinations of 0° and 90°, JRC has a marginal influence on peak stress and elastic modulus. In contrast, as the inclination increases, the peak stress, peak strain, and elastic modulus collectively exhibit an approximate V-shaped trend. The dominant failure mode observed was a mixed splitting-shear mechanism. The number of cracks at final failure increases with higher JRC values under the same joint inclination. As the joint inclination varied, the distributions of global, tensile, and shear cracks all exhibited similar V-shaped trends. Concurrently, the proportions of different microcrack types demonstrated relative stability throughout the failure process, with tensile and shear failures constituting the dominant microscopic mechanisms. Full article

Friction stir spot welding (FSSW) is a novel variant of Friction Stir welding (FSW), developed by Mazda Motors and Kawasaki Heavy Industries to join similar and dissimilar materials in a solid state. It is an economic and environmentally friendly alternative to resistance spot welding (RSW). The FSSW technique, however, includes some structural defects imbedded within the weld joint, such as keyhole formation, hook crack, and bond line oxidation challenging the joint strength. The unique properties of nanomaterials in the reinforcement of metal matrices motivated researchers to enhance the FSSW joints’ strength. Previous studies successfully fabricated nano-reinforced FSSW joints. At different volumetric ratios of nano-reinforcement, nanoparticles may agglomerate due to inefficient stirring of the welding tool pin, forming stress concentration sites and brittle phases, affecting tensile and fatigue strength under static and cyclic loading conditions, respectively. This work investigated how the welding tool pin affects stirring efficiency by controlling the distribution of a nano-reinforcing material within the joint stir zone (SZ), and thus the tensile and fatigue strength of the FSSW joints. Sheets of AA6061-T6 of 1.8 mm thickness were used as a base material. In addition, graphene nanoplatelets (GNPs) with lateral sizes of 1–10 µm and thicknesses of 3–9 nm were used as nano-reinforcements. GNP-reinforced FSSW specimens were prepared and successfully fabricated. Optical microscope (OM) and field emission scanning electron microscope (FE-SEM) methods were employed to visualize the GNPs’ incorporation into the SZs of the FSSW joints. Micrographs of as-welded specimens showed lower formations of scattered, clustered GNPs achieved by the threaded pin tool compared to continuous agglomerations observed when the cylindrical pin tool was used. Tensile test results revealed a significant improvement of about 30% exhibited by the threaded pin tool compared to the cylindrical pin tool, while fatigue test showed an improvement of 46–24% for the low- and high-cycle fatigue, respectively. Full article

The Failure Assessment Diagram (FAD), as a significant method for evaluating the suitability of defective metallic structures, has been subject to considerable debate regarding its applicability in assessing ring welded joints for high-grade steel and large-diameter pipelines. To address this issue, this study first designed and conducted two sets of full-scale pressure-tension tests on large-diameter X80 pipeline ring welded joints, considering factors such as different welding processes, joint configurations, defect dimensions, and locations. Subsequently, three widely adopted failure assessment diagram methodologies—BS 7910, API 579, and API 1104—were selected. Corresponding assessment curves were established based on material performance parameters obtained from the ring weld tests. Finally, predictive outcomes from each assessment method were compared against experimental data to investigate the applicability of failure assessment diagrams for evaluating high-strength, large-diameter, thick-walled ring welds. The research findings indicate that, under the specific material and defect assessment conditions employed in this study, the API 1104 assessment results exhibited significant conservatism (two sets matched). Conversely, the BS 7910 and API 579 assessment results showed a high degree of agreement with the experimental data (eight sets matched), with the BS 7910 assessment providing a relatively higher safety margin compared to API 579. The data from this study provides valuable experimental reference for selecting assessment methods under specific conditions, such as similar materials, defects, and loading patterns. Full article

Fine-grained visual classification requires recognizing subtle inter-class differences under substantial intra-class variation. Ancient mural recognition poses additional challenges: severe degradation and complex backgrounds introduce noise that obscures discriminative features, limited annotated data restricts model training, and dynasty-specific artistic styles manifest as periodic brushwork patterns and compositional structures that are difficult to capture. Existing spatial-domain methods fail to model the frequency characteristics of textures and the cross-region semantic relationships inherent in mural imagery. To address these limitations, we propose a Vision Transformer (ViT) framework which integrates frequency-domain enhancement, explicit token-relation modeling, adaptive multi-focus inference, and discriminative metric supervision. A Frequency Channel Attention (FreqCA) module applies 2D FFT-based channel gating to emphasize discriminative periodic patterns and textures. A Cross-Token Relation Attention (CTRA) module employs joint global and local gates to strengthen semantically related token interactions across distant regions. An Adaptive Omni-Focus (AOF) block partitions tokens into importance groups for multi-head classification, while Complementary Tokens Integration (CTI) fuses class tokens from multiple transformer layers. Finally, Adaptive Margin Contrastive Center Loss (AMCCL) improves intra-class compactness and inter-class separability with margins adapted to class-center similarities. Experiments on CUB-200-2011, Stanford Dogs, and a Dunhuang mural dataset show accuracies of 91.15%, 94.57%, and 94.27%, outperforming the ACC-ViT baseline by 1.35%, 1.63%, and 2.20%, respectively. Full article

This paper proposes a PI-Dual-STGCN fault diagnosis model based on a SHAP-LLM joint explanation framework to address issues such as the lack of transparency in the diagnostic process of deep learning models and the weak interpretability of diagnostic results. PI-Dual-STGCN enhances the interpretability of graph data by introducing physical constraints and constructs a dual-graph architecture based on physical topology graphs and signal similarity graphs. The experimental results show that the dual-graph complementary architecture enhances diagnostic accuracy to 99.22%. Second, a general-purpose SHAP-LLM explanation framework is designed: Explainable AI (XAI) technology is used to analyze the decision logic of the diagnostic model and generate visual explanations, establishing a hierarchical knowledge base that includes performance metrics, explanation reliability, and fault experience. Retrieval-Augmented Generation (RAG) technology is innovatively combined to integrate model performance and Shapley Additive Explanations (SHAP) reliability assessment through the main report prompt, while the sub-report prompt enables detailed fault analysis and repair decision generation. Finally, experiments demonstrate that this approach avoids the uncertainty of directly using large models for fault diagnosis: we delegate all fault diagnosis tasks and core explainability tasks to more mature deep learning algorithms and XAI technology and only leverage the powerful textual reasoning capabilities of large models to process pre-quantified, fact-based textual information (e.g., model performance metrics, SHAP explanation results). This method enhances diagnostic transparency through XAI-generated visual and quantitative explanations of model decision logic while reducing the risk of large model hallucinations by restricting large models to reasoning over grounded, fact-based textual content rather than direct fault diagnosis, providing verifiable intelligent decision support for industrial fault diagnosis. Full article

Tracking knee joint movement during activities of daily living can have the potential to transform the rehabilitation and functional assessment of patients. The present study evaluated the validity of a low-cost, instrumented knee brace to determine whether it was appropriate for the monitoring and quantification of human knee function during five activity-of-daily-living (ADL) tasks including walking, inclined walking, stepping, sitting, and object manipulation. A sensor platform was designed to acquire sagittal plane knee data from 13 healthy participants across five different tasks and compared to gold-standard motion analysis. The brace showed good-to-excellent validity (RMSE: 4.97–8.65°), with differences in knee joint angles and angular velocities noted during various ADLs, specifically during early and late portions of a given movement. The results for instantaneous knee joint angles and angular velocities were very similar to those of the gold-standard system (mean bias: 0.59–9.52°·s⁻¹), which may be applicable to everyday movement tasks, but may preclude analyses at a clinical level. Although the low-cost sensor platform shows promise an effective monitoring tool, it is not ready yet for a clinical application. Full article

Brain magnetic resonance imaging (MRI) is highly susceptible to motion artifacts that degrade fine structural details and undermine quantitative analysis. Conventional U-Net-based deep learning approaches for motion artifact reduction typically operate only in the image domain and are often trained on data with simplified motion patterns, thereby limiting physical plausibility and generalization. We propose Sim-DDNet, a simulation-data-based dual-domain network that combines k-space-based motion simulation with a joint image-k-space reconstruction architecture. Motion-corrupted data were generated from T2-weighted Alzheimer’s Disease Neuroimaging Initiative brain MR scans using a k-space replacement scheme with three to five random rotational and translational events per volume, yielding 69,283 paired samples (49,852/6969/12,462 for training/validation/testing). Sim-DDNet integrates a real-valued U-Net-like image branch and a complex-valued k-space branch using cross attention, FiLM-based feature modulation, soft data consistency, and composite loss comprising L1, structural similarity index measure (SSIM), perceptual, and k-space-weighted terms. On the independent test set, Sim-DDNet achieved a peak signal-to-noise ratio of 31.05 dB, SSIM of 0.85, and gradient magnitude similarity deviation of 0.077, consistently outperforming U-Net and U-Net++ across all three metrics while producing less blurring, fewer residual ghost/streak artifacts, and reduced hallucination of non-existent structures. These results indicate that dual-domain, data-consistency-aware learning, which explicitly exploits k-space information, is a promising approach for physically plausible motion artifact correction in brain MRI. Full article

Background: Accurate microbiological diagnosis of bone and joint infections (BJIs) is frequently hampered by low bacterial load, biofilm formation, and suboptimal tissue processing. This study evaluated the diagnostic performance of mechanical bead-milling using the Ultra-Turrax^® Tube Drive system compared with standard vortex homogenization. Methods: In a prospective cohort of 116 patients undergoing surgery for suspected BJIs, 540 intraoperative samples were processed using both methods. Culture and 16S rRNA PCR results were analyzed using classical and Bayesian statistical approaches. Diagnostic performance was assessed globally and across specimen types and anatomical sites. Results: Ultra-Turrax^® significantly improved sensitivity across all sample types (87.1% vs. 75.2%, p < 0.0001), while maintaining comparable specificity (>99%). Culture positivity increased by 17%, with the greatest gains observed in bone samples and hip prosthesis infections. Quantitative cultures demonstrated a 1.5–2 log₁₀ CFU/mL increase in bacterial yield. In culture-negative specimens, 16S rRNA PCR detection doubled with Ultra-Turrax^® processing (26% vs. 13%, p = 0.04). No increase in contamination was observed. Time to positivity was similar between methods, although Ultra-Turrax^® provided earlier results in 17% of cases. Bayesian modeling confirmed superior sensitivity (posterior probability > 0.995). Conclusions: Ultra-Turrax^® bead-milling markedly enhances microbiological detection in BJIs, particularly in low-biomass and bone-derived specimens. Its simplicity, reproducibility, and compatibility with routine workflows support its integration into diagnostic pathways. This pre-analytical optimization may improve etiological identification and guide more targeted antimicrobial therapy. Full article