Search Results (6,606)

Hungarian teachers’ reform priorities remain insufficiently mapped, despite their central role in shaping feasible, evidence-based educational change. In a cross-sectional study with 1254 kindergarten, primary, and secondary teachers across Hungary (May 2025), we elicited and analyzed open-ended written responses in which participants identified their top three required reforms. Responses were segmented and coded into 18 mutually exclusive categories via a validated codebook, and prevalence was calculated using respondent-normalized weights. We then examined demographic, well-being, and personality correlates of reform priorities using χ² tests, Mann–Whitney tests, and multivariable logistic models with Benjamini–Hochberg false discovery correction. Teachers most frequently prioritized competency development and pedagogical reform, followed by curriculum flexibility and system governance. Reform priorities were not random: female teachers were substantially more likely to prioritize inclusion and SEN support, while male teachers more often prioritized governance and depoliticization; older age predicted governance priorities. Lower educational system satisfaction robustly predicted prioritizing curriculum reform, autonomy, and governance restructuring, and anxiety and depression were positively related to curriculum concerns. Conscientiousness predicted prioritizing salary and material recognition. The results indicate that teachers’ reform demands function as systematic, psychologically grounded signals that can guide more targeted, teacher-centerd educational policy in Hungary. Full article

This article examines whether a high-performing fraud detection model can also meet the demands of auditability, documentation, and regulatory transparency. Using the publicly available European credit card fraud dataset of 284,807 transactions, including 492 fraudulent cases, this study compares weighted logistic regression with XGBoost under severe class imbalance. Model performance is assessed through precision, recall, F1 score, ROC AUC, and precision–recall AUC, with particular attention to alert burden and fraud capture. Results show that XGBoost materially outperforms logistic regression in operational terms. While logistic regression achieves slightly higher recall, XGBoost raises precision from 0.061 to 0.562, improves PR AUC from 0.719 to 0.863, and reduces false positives from 1386 to 67. The PR AUC of 0.863 refers to the cross-validated average reported in the model comparison, while the holdout test result reported later in this paper is 0.852. It cuts the review queue from 1476 alerts to 153 while still identifying 86 of 98 fraud cases in the test set. Explainability is then introduced through SHAP, which provides both global feature attribution and transaction-level reasoning. The findings show that SHAP makes the boosted model readable at the level of both overall model behaviour and individual fraud flags, thereby supporting audit review, model validation, and regulatory scrutiny. The article argues that the combination of XGBoost and SHAP offers a stronger fit for auditing than either a weaker but transparent linear model or a stronger opaque classifier. One limit remains, since the dataset contains anonymised principal components rather than original business variables, which restricts semantic interpretation. Even so, the workflow provides a practical bridge between predictive fraud analytics and the demands of explainable, reviewable, and accountable AI in auditing. Full article

Background/Objectives: The interplay between cardiovascular risk factors and brain aging remains incompletely understood. We aimed to investigate the comparative associations of coronary artery calcium (CAC) and low-density lipoprotein cholesterol (LDL-C) with MRI-derived volumetric measures of the brain. Methods: In this retrospective, single-center, cross-sectional study, 84 participants who underwent coronary computed tomography for CAC scoring and brain magnetic resonance imaging within 90 days were included; LDL-C levels were available in 69 participants for LDL-based analyses. Brain volumetric measures were obtained using the automated lesionBrain pipeline within the volBrain platform, which performs fully automated tissue segmentation and lesion quantification based on multi-atlas and patch-based approaches. Associations were evaluated using Spearman’s correlation with false discovery rate correction and hierarchical multivariable regression, supported by bootstrap validation and post hoc power analysis. The cohort had a mean age of 58.0 ± 13.0 years (range 19–78) and was derived from routine clinical imaging. Results: LDL-C was positively associated with abnormal white matter volume (ρ = 0.334, p = 0.005), although this did not remain statistically significant after FDR correction (pFDR = 0.090). In fully adjusted models, LDL-C remained the only independent predictor (β = 0.006, 95% CI: 0.002–0.010, p = 0.007; standardized β = 0.225; partial R² = 11.7%), corresponding to a 6.2% increase in abnormal white matter volume per 10 mg/dL increase (derived from log-transformed models). CAC showed only a marginal association (p = 0.059). Post hoc power analysis demonstrated adequate power for LDL-C but insufficient power for CAC. Neither marker was associated with gray matter volume. Conclusions: In this cross-sectional cohort, higher LDL-C was independently associated with greater abnormal white matter volume after adjustment for cardiovascular risk factors, statin use, and CAC. No CAC–brain association was detected in this cohort, but limited statistical power means that small CAC effects cannot be excluded. These findings should be interpreted as associative rather than causal or mechanistic. Full article

Plant disease detection in natural environments is significantly challenged by variations in lesion scales and interference from complicated background clutter. Nevertheless, current models often remain limited in effectively capturing multi-scale features and mitigating background interference simultaneously. To tackle these challenges, we present TGL-YOLO, an improved detection network built on the YOLO11 framework. Methodologically, we introduce the Tri-Scale Dynamic Block (TSDBlock) to adaptively extract fine-grained features across highly variable lesion sizes. Furthermore, a Gated Pyramid Spatial Transformer (GPST) is designed to fuse cross-scale features and suppress background interference, while a Large Separable Pyramid Attention (LSPA) module expands the spatial receptive field to capture global context. Experimental results on two public datasets show that TGL-YOLO demonstrates improved performance over the YOLO11s baseline. On the PlantDoc dataset, it improves mAP50 and mAP50:95 by 4.7% and 3.7%, reaching 0.591 and 0.449, respectively. On the FieldPlant dataset, it reaches 0.793 and 0.608, yielding improvements of 2.3% and 1.9%. The proposed method demonstrates the capability to reduce missed detections and false positives caused by multi-scale lesions and environmental noise, providing a competitive and computationally viable solution for agricultural disease monitoring in natural environments. Full article

Ship target detection is a prerequisite for achieving automated monitoring in ship detection systems. To address the challenge of accurately detecting ship targets in complex water environments, this study proposes a ship target detection method based on an improved YOLOv11 framework. To enhance the model’s ability to perceive and fuse features across multiple scales and in complex backgrounds, an Iterative Attention Feature Fusion (iAFF) module and a Biformer module are integrated at the end of the backbone network. The iAFF module iteratively optimizes multi-scale features through a two-stage attention mechanism, effectively focusing on key target regions, thereby improving the model’s detection capability for small, medium-sized, and occluded ships. The Biformer module leverages its innovative Bi-level Routing Attention (BRA) mechanism to enhance the modeling of global semantic information while reducing computational complexity, mitigating false detections caused by occlusions among ship targets, and consequently improving detection precision. This study employs the Minimum Point Distance Intersection over Union (MPDIoU) loss function, which more comprehensively measures the similarity between predicted and ground-truth bounding boxes by optimizing the distances of their key geometric points, effectively enhancing the accuracy of bounding box regression. Experimental results show that the proposed model achieved 93.96% mAP, 92.93% recall, and 94.97% precision on a self-built ship dataset, surpassing mainstream detection algorithms including YOLOv11 in multiple metrics. The model has only 2.90 M parameters, achieving a good balance between accuracy and efficiency. This provides an accurate and efficient solution for intelligent ship supervision. Full article

Infrared small-target segmentation (IRSTS) is crucial for a wide range of applications, including maritime search-and-rescue operations and intelligent traffic surveillance. However, current deep learning methods struggle with dynamic scale variations in infrared small targets, resulting in false detections and missed detections, alongside inadequate core localization accuracy. To address these challenges, we propose an infrared small-target segmentation framework founded on morphological attention and an energy core loss function, IRSTS_Unet. Specifically, we design a Dynamic Shape-adaptive Deformable Attention Module (DSDAM), which achieves parameterized feature extraction via “initial localization–offset deformation–precise sampling”. This approach enables the network to differentially focus on target cores and background cues to suppress clutter. To improve the efficiency of multi-scale feature aggregation, we embed the DSDAM within both the feature extraction and cross-layer fusion stages. Furthermore, we formulate a Core Energy-aware Core-Priority loss (CECP-Loss) function that incorporates the energy prior distribution of small targets, effectively counteracting the “core dilution” phenomenon endemic to conventional loss functions. Through extensive experiments on multiple public datasets, we demonstrate that IRSTS_U-Net outperforms state-of-the-art approaches in terms of both detection accuracy and robustness. Full article

The Low-Strain Pile Integrity Test (LSPIT), standardized in ASTM D5882, is widely used as a rapid and economical non-destructive technique for assessing pile continuity in deep foundation systems. However, interpretation of LSPIT reflectograms remains strongly dependent on expert judgment and is influenced by soil–pile interaction effects such as damping and radiation losses, which can alter waveform morphology and confound automated defect screening. This study proposes a soil-aware deep learning framework that combines image-based reflectogram features with categorical geotechnical context describing the dominant soil regime at the measurement site. Reflectogram images are processed with a pretrained ConvNeXt-Large backbone, while soil information derived from Unified Soil Classification System (USCS) logs is represented as a categorical auxiliary input and mapped to a learnable embedding. The resulting multimodal design conditions waveform interpretation based on site context rather than relying on signal morphology alone. The framework is examined on an assembled benchmark of 510 expert-labeled reflectograms (404 intact and 106 defective), including a nine-site subset of 182 field records with explicit soil annotations. On the assembled benchmark, the model yields 99.41% accuracy and a weighted F1-score of 0.9941; on the nine-site subset, the observed accuracy is 99.45% with zero missed defective cases. Balanced accuracy, specificity, missed-detection rate, false-alarm rate, and confidence intervals are additionally reported to better align the evaluation with engineering screening practice. The study also states the current limits of the evidence base, including partial soil annotation, dominant-soil simplification, restricted soil coverage, and the absence of leave-site-out and interpretability-focused validation. Overall, the results support soil-aware multimodal learning as a promising proof-of-concept direction for more context-aware automated LSPIT interpretation, while also identifying the validation steps still required for broad field deployment. Full article

In response to the subjective issues, inconsistent quality standards, high labor intensity and low sorting efficiency during the drying process of green pepper, an improved YOLOv11 algorithm was proposed for quality detection. A multi-scale edge enhancement module (MEEM) is introduced into the backbone network, replacing the original basic C3K2 module with C3K2-MEEM to enhance the extraction of detailed features in images of dried green Sichuan pepper and prevent missed detections, false detections, and boundary confusion. The LRSA module is integrated into the 10th layer of the backbone network to improve the clarity of the tumor-like texture of the Sichuan pepper and reduce the influence of impurities, automatically allocating attention based on feature similarity to preserve local information. In the neck layer, the DPCF module is added to the FPN+PAN feature fusion stage to achieve multi-scale feature collaboration, meeting the detection requirements of dried green Sichuan pepper. The results show that the accuracy recall rate, mean average precision, and model size of the improved MLD-YOLOv11 algorithm are 92.1%, 96.6%, 95.6%, and 11.06 MB, respectively. Compared with the training results of the original YOLOv11 model, the average accuracy of the improved model has increased by 2.2 percentage points, and GFLOPs have definitely decreased by 2 G, with parameter reduction of approximately 3.10%. Compared with other mainstream models, the MLD-YOLOv11 model has significant advantages in terms of mean average precision, model size, and floating point operations per second, making it more suitable for industrial applications and providing an efficient, accurate, and lightweight solution for the quality detection of dried green Sichuan pepper. Full article

Background/Objectives: Although clinically distinct, bipolar disorder (BP), schizophrenia (SZ), major depressive disorder (MDD), and social anxiety disorder (SAD) share fundamental biology. We mapped these transdiagnostic systemic mechanisms. Methods: Weighted Gene Co-Expression Network Analysis (WGCNA) of peripheral blood RNA-Seq datasets evaluated module preservation, hub gene disruption, and microRNA (miRNA) networks. Results: Seven modules showed robust cross-disease preservation. Overall, 56 of 105 candidate hub genes exhibited altered expression, with 22 passing the false discovery rate (FDR) correction. Hubs like IL1B, TLR2, and MMP9 dominated networks linked to altered inflammatory signaling and structural remodeling. Downregulated ribosomal hubs characterized systemic metabolic stress. Discussion: These signatures capture extensive systemic dysregulation. Inflammation and metabolic shifts correlate strongly with pathways regulating chronic neuroinflammation, epigenetic control, and dendritic pruning. Computational models suggest these cascades evade miRNA controls, potentially compromising structural neural plasticity. Conclusions: This shared transcriptomic architecture challenges rigid diagnostic boundaries. Identifying systemic immune dysregulation and translational alterations as core pathogenic denominators provides a rationale for transdiagnostic therapies targeting upstream systemic networks to mitigate neural vulnerabilities. Full article

Hail events commonly affect the Western part of Catalonia, producing damage mainly in the agricultural sector. Comparison of the weather radar data with hail pad registers at ground level allows for the diagnosis of hail severity. However, limitations using individual radar fields have led to the use of quantiles of the vertical profiles of reflectivity for a period between 12 min before and after a hailfall. These profiles combine all radar parameters, and are less sensitive to radar functioning anomalies and hailfall nature. The explored dataset was divided into severe and non-severe registers, with two subsets: one larger (90% of cases) for modeling and the second one for validating the results. Results indicate a better estimation of severe hail, but the number of false alarms with non-severe cases was still high. In consequence, future work should focus on minimizing false alarms using more restrictive profile groups. The purpose of the study is the application of a real-time tool for improving surveillance tasks which provides better discrimination between severe and non-severe hail occurrences. Full article

Objectives: This study evaluated the diagnostic performance of the lateral flow immunochromatographic assay (RESIST-ACINETO, Coris BioConcept) for the rapid detection of the major carbapenemases in Acinetobacter baumannii. Methods: Blood culture isolates collected between 2014 and 2024 with meropenem MIC ≥ 4 mg/L were retrieved, re-identified by MALDI-TOF MS, and susceptibility was confirmed by broth microdilution. Carbapenemase genes (bla_OXA-23, bla_OXA-40, bla_NDM) were detected using multiplex PCR, which served as the reference standard. All isolates were tested using the RESIST ACINETO assay, and diagnostic accuracy parameters were calculated. Results: A total of 114 isolates were recovered and confirmed as A. baumannii. Among 93 carbapenem-non-susceptible isolates, 97.8% (91/93) were correctly identified by the assay. The test showed 99.1% sensitivity and 99.1% specificity, with most positive results appearing within 3–10 min. Two discrepant results were observed (one false positive, one false negative), while all meropenem-susceptible isolates tested negative. Conclusions: The RESIST ACINETO assay provides rapid, accurate detection of carbapenemases in A. baumannii, significantly reducing turnaround time compared with conventional workflows. Its performance supports integration into routine diagnostics to enhance timely resistance confirmation and infection-control interventions. Full article

Background: Untargeted proteomics enables quantitative host cell protein (HCP) determination in biotherapeutics, yet no workflow has been validated under ICH Q2(R2) for regulated quality control. Methods: A prospective total-error (TE) validation of label-free ddaPASEF proteomics was performed. A stable isotope-labeled whole-proteome standard was spiked into NISTmAb at seven levels (20–80 ng) and analyzed in four independent assays (198 injections), supporting one-way random-effects ANOVA with Welch–Satterthwaite adjustment. Peptide-level identification error was evaluated by dual entrapment. Results: Empirical false-discovery proportions were below 1% at q = 0.01. Weighted least-squares regression (R² = 0.993) confirmed stable proportional compression with 81–85% recovery. Repeatability dominated the variance structure (median CV 2.7%); intermediate precision SD ranged from 0.69% to 3.81%. Both 95% β-expectation and 95/95 content tolerance intervals were contained within ±30% at all levels, defining a validated range of 20–80 ng. Abundance-stratified TE profiling revealed concentration-dependent calibration heterogeneity, with stratum-specific intervals within ±35% defining an abundance-aware LLOQ of 3.6 ppm (P95 = 3.87 ppm). Robustness under independent search software (FragPipe v24.0, CCC = 0.998) and cross-platform acquisition (Astral, CCC = 0.980) remained within ±30% limits. Conclusions: This constitutes the first prospective ICH Q2(R2)-aligned validation of untargeted proteomics for HCP quantification, with a transferable statistical framework for high-dimensional analytical methods. Full article

Traffic accident detection is a critical component of intelligent transportation systems (ITS), enabling timely incident response and traffic management. While most existing approaches rely on temporal information from video sequences, such methods are not always applicable in resource-constrained surveillance environments. This study investigates the feasibility of detecting traffic accidents from single static images by formulating the task as a binary classification problem. Representative architectures, including Vision Transformer (ViT), Swin Transformer, and ResNet-50, are systematically evaluated on the Car Crash Dataset (CCD) under multiple training configurations. To assess generalization capability, cross-domain evaluation is conducted using an external crash video dataset (ECVD) constructed to approximate real-world deployment conditions. Experimental results show that all models achieve strong performance under in-domain evaluation. However, cross-domain testing reveals substantial performance degradation, particularly in recall, indicating limited generalization capability under domain shift. Qualitative analysis further shows that missed detections are associated with weak visual cues, occlusion, and complex traffic environments, while false positives are caused by visually ambiguous patterns resembling accident scenarios. Unlike prior studies that primarily report performance improvements, this work provides empirical evidence that model behavior in static-image-based accident detection is governed by dataset composition rather than architectural design. Therefore, static-image-based accident detection should be interpreted as a coarse-level screening tool rather than a fully reliable decision-making system. This study highlights the importance of data-centric design and cross-domain evaluation for improving real-world applicability. Full article

Background: Trace elements function as essential micronutrients involved in oxidative balance, mitochondrial activity, and cardiovascular metabolism. Cigarette smoking represents a significant source of toxic metals and may disrupt systemic trace element homeostasis. Alterations in micronutrient and metal balance may contribute to oxidative stress, endothelial dysfunction, and myocardial remodeling, which are central mechanisms in the pathogenesis of heart failure with preserved ejection fraction (HFpEF). This study aimed to investigate whether smokers with HFpEF exhibit distinct hair trace element profiles compared with smokers without HFpEF. Methods: In this prospective pilot study, scalp hair samples were collected from adults undergoing clinical evaluation for suspected cardiovascular disease. Trace element concentrations were determined using inductively coupled plasma mass spectrometry (ICP-MS). Participants were first stratified according to smoking status and subsequently, within the smoker subgroup, according to HFpEF diagnosis based on the Heart Failure Association Pre-test assessment, Echocardiography and natriuretic peptide score (HFA-PEFF) algorithm. Differences in trace element concentrations were analyzed using appropriate statistical tests, with multiple-comparison correction using the Benjamini–Hochberg false discovery rate (FDR). Active smoking was defined as ≥10 cigarettes per day for at least 1 year, and cumulative exposure was quantified in pack-years. Results: Fifty-eight participants were included, including 27 active smokers. In unadjusted analyses, several trace elements differed between smokers with HFpEF and those without HFpEF, including vanadium, lithium, aluminum, and copper. However, after FDR correction, only copper remained significantly elevated in smokers with HFpEF (q = 0.004). Hair copper concentrations were markedly higher in the HFpEF group compared with smokers without HFpEF. These differences were observed alongside echocardiographic features consistent with diastolic dysfunction and structural cardiac remodeling. Conclusions: In this hypothesis-generating pilot study, smokers with HFpEF demonstrated elevated hair copper concentrations, suggesting disturbances in trace element and micronutrient homeostasis. Altered copper metabolism may reflect oxidative stress-related cardiometabolic remodeling associated with HFpEF. These findings raise the hypothesis that cardiometabolic phenotype, rather than smoking exposure alone, may modulate trace element homeostasis in HFpEF; however, causal relationships cannot be established. Full article

Electroluminescence (EL) imaging is widely used for photovoltaic (PV) module inspection, yet EL defect detection remains challenging due to the need for high-resolution inputs, low-contrast defects, and strong structured background patterns. To address these issues, we propose MSW-Mamba-Det, an end-to-end defect detection framework built on RT-DETR, comprising three components. (1) MSW-Mamba, a multi-scale windowed state-space module, adopts a Local/Stripe/Grid architecture to jointly model fine details and long-range dependencies; the Stripe branch strengthens directional continuity for elongated defects, while the Grid branch introduces coarse global context to improve cross-region consistency. Saliency- and gradient-guided gating is further used to suppress background-induced false responses. (2) DetailAware compensates for detail attenuation by restoring high-frequency textures and edges through multi-scale local enhancement, and applies pixel-wise adaptive gating to integrate global semantics and mitigate smoothing effects in deep representations. (3) PAFB (Pyramid Attention Fusion Block) aligns adjacent-scale features and improves multi-scale fusion, enhancing localization stability across defect sizes. Experiments on two public EL datasets show that MSW-Mamba-Det achieves $A{P}_{50:95}$ of 60.4% on PV-Multi-Defect-main and 68.0% on PVEL-AD, improving over RT-DETR by 2.5 points (from 57.9% to 60.4%) and 2.2 points (from 65.8% to 68.0%), respectively. MSW-Mamba-Det also outperforms 12 representative baselines, including CNN-, Transformer-, and recent YOLO-based models, in $A{P}_{50:95}$ on both datasets, with particularly strong performance on medium and large defects. These results demonstrate the effectiveness of the proposed modules for robust PV EL defect inspection under low-contrast and structured-background conditions. Full article