Search Results (9,954)

Permanent magnet synchronous motors (PMSMs) have been widely used in new-energy vehicles and industrial servo systems. However, demagnetization faults (DMFs) can lead to severe issues, including torque ripple and magnetic field distortion. This paper proposes an intelligent diagnostic approach for DMFs based on stator tooth flux (STF). A mathematical model of STF is formulated, and the magnetic flux change is measured using multiple sets of anti-series-connected detection coils (DCs). By combining finite element simulation with signal processing technology, we establish a comprehensive diagnostic system covering fault feature extraction, fault location identification, and severity assessment is established. The proposed method employs wavelet transform (WT) to extract time-frequency features of voltage signals and combines it with a convolutional neural network (CNN) to form the WT-CNN intelligent diagnosis model. Based on the extracted voltage signal features, the method achieves intelligent identification and visual localization of DMFs. Simulation results show that the proposed method achieves an accuracy above 80% for fault location identification (defined as sample-level multi-label classification accuracy across 12 PMs) and above 85% for demagnetization severity estimation (defined as classification accuracy across 9 severity degrees from 10% to 90%). These results provide an effective technical foundation for motor condition monitoring and fault early warning in simulation environments. Full article

Media coverage of disasters frequently frames self-interested behavior in contrast to collective responsibility and coordinated response. This study aims to explore how such behavior is emotionally constructed in disaster-related media, using a carefully selected corpus of 12 text-centered news articles focusing on selfish behavior. The analysis combines transformer-based sentence-level emotion classification using the tweetnlp RoBERTa model, which predicts 11 emotion categories, with Latent Dirichlet Allocation topic modeling across single-sentence and three-sentence windows in a small purposively selected corpus. Emotion–topic relationships are quantified by weighting emotion probabilities by topic distributions and visualized using bar charts, network graphs, and heatmaps. The findings suggest that fear and disgust dominate portrayals of self-interested behavior, while anticipation appears in projections of harm and anger is linked to inequality and institutional accountability. Two discursive configurations emerge: Responsibility Across Individuals and Institutions, emphasizing public accountability and authority, and Collective Fear and Self-Protective Practices, reflecting affect-driven responses under uncertainty. Although negative emotions predominate, optimism appears conditionally, signaling coordination and recovery. Overall, disaster reporting constructs selfishness through integrated emotional–semantic patterns that position individual actions within broader social risk and collective responsibility. Full article

Skin cancer represents a significant public health challenge, necessitating early detection and timely treatment for optimal management. Timely and accurate evaluation of skin lesions is crucial, as delays can lead to more severe outcomes. However, identifying skin lesions accurately can be challenging due to differences in color, shape, and the various types of imaging equipment used for diagnosis. While recent studies have demonstrated the potential of ensemble convolutional neural networks (CNNs) for early diagnosis of skin disorders, these models are often too large and inefficient for processing contextual information. Although lightweight networks like MobileNetV3 and EfficientNet have been developed to reduce parameters and enable deep neural networks on mobile devices, their performance is limited by inadequate feature representation depth. To mitigate these limitations, we propose a new hybrid attention dual-stream deep learning model for skin lesion detection. Our model uses one training process to preprocess the images and splits the task into two branches. Each branch extracts different features using multi-stage and multi-branch attention techniques, improving the model’s ability to detect skin lesions accurately. The first branch processes the original image using a convolutional layer integrated with three novel attention modules: Enhanced Separable Depthwise Convolution (SCAttn), stage attention, and branch attention. The second branch utilizes Contrast Limited Adaptive Histogram Equalization (CLAHE) to enhance the input image, improving local contrast and revealing finer details. The integration of CLAHE with SCAttn modules leverages enhanced local contrast to capture more nuanced features while maintaining computational efficiency. A classification module receives the concatenated hierarchical characteristics that were taken from both branches. Utilizing the PAD2020 and ISIC 2019 datasets, we assessed the proposed model and obtained an accuracy rate of 98.59% for PAD2020, surpassing the state-of-the-art performance by 2%, and stable performance accuracy for the ISIC 2019 dataset. This illustrates how well the model can integrate several attention mechanisms and feature enhancement methods, providing a reliable and effective means of detecting skin cancer. Full article

The increasing reliance on smart grid communication systems has significantly raised the demand for robust cybersecurity measures to defend against advanced threats. This paper proposes a two-tier classification framework to enhance the detection and diagnosis of man-in-the-middle attacks within smart grid communication protocols. Initially, the model detects the presence of an attack and then identifies the specific type of man-in-the-middle attack through subsequent inferences. To achieve this, the “Man-in-the-Middle Attacks Targeting Modbus TCP/IP and MMS Protocols in the Smart Grid” dataset was carefully preprocessed and analyzed to better understand the underlying hidden characteristics. This understanding, coupled with existing works on fault detection and diagnosis, facilitated the engineering of new features from the original dataset. Four classifiers were employed in each tier: Random Forest, XGBoost, LightGBM, and CatBoost. The first tier exhibited exceptional performance, with the CatBoost framework achieving 99.6% accuracy. The second tier also demonstrated strong results, with the same model achieving 99.1% accuracy. Systematic model explainability was conducted using SHapley Additive exPlanations for both tiers and revealed that the highest accuracy was achieved using five features for the first and six for the second. The average inference time was approximately 4.76 milliseconds. The proposed framework is accurate, fast, interpretable, lightweight, and well-optimized for direct implementation in smart grid systems to detect and diagnose man-in-the-middle attacks. Full article

Oceanfront relief varies along coastlines and serves as the first barrier to wave and surge damage. However, forecasted increases in storm frequency and sea levels are anticipated to enhance coastal erosion, potentially weakening this protection. The land–sea transition is variable along the New England coast, USA, and this variability has produced a range of coastal morphologies that can vary over short distances. It is important to track the beach transition zone to better understand transformations of the system and related hazard risks. A combination of field and computer-based methods was used to evaluate the beach transition zone of southwestern Rhode Island to determine alongshore variability and dynamics. More specifically, a decadal-scale study was conducted to examine changes in morphology from 2011 to 2022, and a short-term study at South Kingstown Town Beach examined changes from November 2023 to January 2024 using time-series drone-derived elevations. Classification of over 500 cross-shore transects illustrated the dominance of sedimentary shorelines, with smaller areas of rocky outcrops and hardening. Analysis of four different years (2011, 2014, 2018, and 2022) determined that beaches with dune morphology were the most common type of transition zone (41–47% of the transects) and transects with a high bank upland were the next most frequent class (34–41%). Following Hurricane Sandy in 2012, a 6% decrease in the number of dune-classified transects was measured; however, one-third of those recovered dune morphology by 2022. The greatest beach transformations over the short-term study occurred in response to strong storms in the 2023–2024 winter season, during which lateral beach movement (erosion) exceeded 15 m in portions of South Kingstown Town Beach. Dune erosion was accompanied by overwash flooding and deposition, and the area remained low-lying and thus vulnerable to future impacts. The beach transition zone classification and insights from this research will be informative for future planning by coastal communities by determining at-risk shorelines based on underlying geology and the stability of morphological features. Full article

Background: Topological Data Analysis (TDA) captures multi-scale geometric features of data as persistence diagrams, yet no principled information-theoretic framework quantifies how much information those features carry, how efficiently they compress, or when they are informationally irreducible. Methods: We construct a measure-theoretic probability space over persistence diagram space using a Poisson-process reference measure, and define topological entropy (H-T), topological mutual information (I-T), and a topological rate–distortion function as the core objects of a new theory. Results: Four theorems with full proofs establish finite stability, axiomatic uniqueness, a Topological Data Processing Inequality, and a Rate–Distortion Theorem with explicit Poisson-model closed-form formula. A Renyi generalization of topological entropy is also established. Computational and practical implementation aspects—including finite-sample estimation, multi-parameter extension, and algorithmic realization—are addressed inline throughout the paper. Conclusions: This framework provides a rigorous measure-theoretic information-theoretic foundation for persistent homology, demonstrated on simulated brain connectivity and point cloud data, with applications to threshold selection, genomic classification bounds, and compressed sensing. Full article

Diffuse midline gliomas (DMGs) are rare but highly aggressive central nervous system (CNS) tumors that can present in both pediatric and adult populations. These tumors were redefined in the 2016 WHO classification of CNS tumors based on integrated histopathological and molecular features, and were initially designated as “DMG, H3 K27M-mutant”. In the 2021 WHO update, DMGs were incorporated into the newly defined category of primarily pediatric-type diffuse high-grade gliomas, and nomenclature was changed to “DMG, H3 K27-altered” to encompass additional molecular drivers beyond the canonical H3 K27M mutation. Clinically, DMGs arise as expansile, infiltrating tumors within midline structures and may present as non-enhancing or enhancing lesions on imaging. Diagnosis is based on neuroimaging and molecular confirmation by immunohistochemistry or sequencing when tissue is available. DMGs are categorized as WHO grade 4 malignant tumors due to their aggressive biology leading to rapid and infiltrative growth. Owing to their deep and midline location, surgical resection is typically not feasible. Radiation therapy is the backbone of treatment, but there is no standard regimen of chemotherapy that has demonstrated durable efficacy. Recent progress in therapeutic approaches has led to a major breakthrough on 6 August 2025 when the U.S. Food and Drug Administration granted the accelerated approval of dordaviprone (ONC201), marking it as the first systemic therapy for progressive DMG harboring H3 K27M mutation. Other novel approaches, including chimeric antigen receptor (CAR) T-cell directed therapies and convection-enhanced delivery, are actively under investigation. We aim to comprehensively review DMGs, including the recent insights into their biology, the evolving therapeutic landscape, and the opportunities to fuel this new momentum against one of the most formidable gliomas. Full article

The exponential rise of e-commerce platforms has resulted in a dramatic increase in online reviews, which creates a challenge in distinguishing fake reviews that erode consumer confidence and harm commerce ecosystems. Traditional approaches for fake review detection employ computationally expensive deep learning networks which are resource-intensive and difficult to use in practice. In this paper, we describe AdaptiveNet, a new lightweight neural architecture that achieves fake review detection with much lower computational resources while maintaining a higher detection and classification precision. The model proposed in this paper is based on three original innovations: a Multi-Scale Semantic Fusion (MSSF) layer for hierarchical feature extraction, Dynamic Attention Scaling (DAS) with complexity measure attention, and Adaptive Parameter Sharing (APS) context-gated networks. With thorough evaluation on Amazon, Yelp, and TripAdvisor datasets of reviews totalling 1.2 million reviews, AdaptiveNet attains 94.8% accuracy while achieving 65% computational overhead in comparison to traditional models. The architecture outperformed all other state-of-the-art models, BERT-base (92.1%), RoBERTa (91.8%), and other more recent efficient models, requiring 70% lower parameters and 60% lower energy consumption. This work markedly advances the other efficient deep learning architectures for text classification and allows for the practical implementation of fake review detection systems in resource-limited settings as process innovation. Full article

The widespread dissemination of fake news across different domains exerts a negative impact on social order. Current fake news detection models face two major challenges. First, the issue of domain shift constrains the generalization capability of models in cross-domain scenarios. Second, general neural networks struggle to extract features between distant words in text, resulting in poor quality of original features and adversely affecting the final detection outcomes. In response to the aforementioned issues, this paper proposes a multi-domain fake news detection framework based on multi-view hybrid attention enhancement. Firstly, superior original feature extraction is achieved through Recurrent Convolutional Neural Networks (RCNN) and Bidirectional Long Short-Term Memory (BiLSTM). Secondly, a hybrid attention mechanism is established between features and domains across multiple views—including news semantics, sentiment, and style—thereby forming domain-specific memory. This enables the model to achieve more precise classification of news within specific, subdivided domains. Finally, experiments conducted on the public dataset Weibo21 demonstrate that the proposed method attains F1 scores of 93.26% and 85.31% on Chinese and English datasets. Full article

Background: Postoperative delirium (POD) is a common complication in geriatric patients. This prospective cohort study evaluated a possible influence of intraoperative positioning on the occurrence of POD, as intraoperative prone positioning could affect cerebral perfusion. Methods: We included 760 patients of ³60 years scheduled for elective surgery in prone or supine positions. The primary outcome was POD incidence on the first five days after surgery, assessed via 3D-Confusion Assessment Method (3-D CAM) or Confusion Assessment Method for Intensive Care Units (CAM-ICU). Preoperative assessments included the American Society of Anesthesiologists (ASA) and New York Heart Association (NYHA) classifications as well as short screenings for the cognitive (modified Montreal Cognitive Assessment (MoCA)) and self-care status of the patient. Secondary outcomes were length of hospital stay (LOS) and mortality rates. Results: Postoperative delirium rates were similar in prone and supine patients (7.6% vs. 5.5%; p = 0.31), and logistic regression analysis revealed no association of intraoperative prone positioning with POD (odds ratio 1.42 (95% CI 0.68–2.92; p = 0.342)). The overall incidence of POD was 6.1% and was associated with older age (81.5 (CI 76.2–84.8) vs. 72.0 (CI 67.0–79.0) years; p < 0.01), higher ASA and NHYA classifications, lower preoperative modified MoCA, reduced independence in self-care (p < 0.001, respectively), and longer incision-to-suture times (107.0 (CI 73.0–173.0) vs. 85.0 (CI 60.0–130.0) minutes; p < 0.01). Postoperative delirium resulted in longer LOS (14.5 (CI 9.0–27.0) vs. 7.0 (CI 4.0–9.0) days; p < 0.001), and increased mortality (13.0% vs. 1.7%; p < 0.001). Conclusions: Intraoperative prone positioning was not associated with POD in patients aged 60 years or older (OR 1.42; CI 0.68–2.92; p < 0.340), and LOS and mortality as secondary outcome parameters were also similar in patients after prone and supine surgery. Future studies assessing additional and possible confounding factors and intraoperative systemic and regional hemodynamics and oxygenation are needed to verify this result and to evaluate cerebral hypoperfusion as a possible mechanism of POD. Full article

Urban remote sensing provides an efficient and accessible way to monitor and assess the urban environment. However, the difficulty in classifying bare soil and built-up land is exacerbated in desert landscapes, due to the spectral confusion of bare soil and impervious surfaces. Therefore, urban remote sensing research in desert environments employs complex and time-consuming classification techniques, which cause difficulties in reliability when transferring these methods to other desert cities. This paper describes two new index-based approaches that can successfully detect and classify urban areas without the disruption of bare soil influences in desert environments using Landsat 8–9 satellite imagery. They are called the desert urban landscape index (DULI) and the isoline impervious surface index (IISI). The desert cities of Phoenix, Ciudad Juárez, and Riyadh were used as study areas for the development of these indices. The two proposed indices outperformed the dry built-up index (DBI), with overall accuracy rates of 85% in Phoenix using DULI, 87% in Ciudad Juárez using DULI, and 90% in Riyadh using IISI. DULI also demonstrates the ability to suppress landscape features such as bare soil, mountains, and canyons. Full article

Event detection is crucial for disaster response, public safety, and trend analysis, enabling real-time identification of critical events. Social media platforms provide a vast content source, offering timely and diverse event coverage compared to traditional news reports. However, challenges arise due to the informal and noisy nature of the text, along with the limited availability of ground truth data for training models. This study introduces SOCIAL (Social Media Event Classification using Integrated Artificial Learning and Natural Language Processing), a mathematically grounded framework for real-time social media event detection. SOCIAL integrates a formal representation of social media text with a customized CNN–LSTM architecture, combining convolutional operations for local feature extraction with sequential modeling to capture temporal dependencies, thereby enhancing classification accuracy. Generative AI is employed to create synthetic event-related samples, addressing data scarcity and ensuring a balanced dataset, while the design incorporates quantitative principles to guide embedding selection and model optimization. This study systematically evaluates six experimental configurations with TF-IDF and Word2Vec embeddings. The TF-IDF-based CNN–LSTM model achieved top performance with 98.59% accuracy, 98.13% precision, 99.06% recall, and 0.9719 MCC. Additionally, the $F_{0.5}$, $F_1$, and $F_2$ scores were 98.31%, 98.59%, and 98.87%, respectively, confirming the model’s strong predictive capabilities. TF-IDF integration enhanced event-specific term recognition, reducing misclassifications and improving reliability. These results demonstrate that SOCIAL is not only a fast, accurate, and scalable tool for crisis event detection, but also a formally principled framework for modeling and analyzing social media signals. Full article

A homologous classification for vespid venoms is missing. This study compared Polistes dominula and Vespula spp. venoms to evaluate their homology level. P. dominula and Vespula spp. extracts, including V. germanica, V. maculifrons, V. pensylvanica, V. alascensis, and V. squamosa in equal proportions, were generated from venom sacs and were subjected to sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) and Western blot using Vespula-positive sera. Bands described as allergenic were excised and sequenced through Liquid Chromatography–Mass Spectrometry tandem analysis (LC-MS/MS) to confirm their identity. Phospholipase (group 1) and hyaluronidase (group 2) enzymatic activities were measured. Group 1 and 5 3-D structures and sequence identity were analyzed in silico. The results showed that the P. dominula and Vespula spp. venom extracts exhibit similar protein profiles and comparable allergen composition, with phospholipase and hyaluronidase activities. The structures of Pol d 1 and Ves v 1 and Pol d 5 and Ves v 5 were highly similar, and the identity levels were high across and within the Polistes and Vespula genera (≥50%). These results suggest the inclusion of venoms from Polistes and Vespula genera as candidates to create a new homologous group for wasp venoms and indicate that the currently described homologous groups require revision. Full article

Unmanned aerial vehicles (UAVs) have become indispensable tools in precision agriculture and plant phenotyping, enabling the rapid, non-destructive assessment of crop traits across space and time. Equipped with RGB, multispectral, thermal, and other sensors, UAVs provide detailed information on canopy structure, physiology, and stress responses that can guide management decisions and accelerate breeding programs. Despite these advances, the downstream processing of UAV imagery remains technically demanding. Converting orthomosaics into standardized, biologically meaningful data often requires a combination of photogrammetry, geospatial analysis, and custom scripting, which can limit reproducibility and accessibility across research groups. We present drone2report, an open-source python-based software that processes orthomosaics from UAV flights to generate vegetation indices, summary statistics, derived subimages, and text (html) reports, supporting both research and applied crop breeding needs. Alongside the basic structure and functioning of drone2report, we also present five case studies that illustrate practical applications common in UAV-/drone-phenotyping of plants: (i) thresholding to remove background noise and highlight regions of interest; (ii) monitoring plant phenotypes over time; (iii) extracting information on plant height to detect events like lodging or the falling over of spikes; (iv) integrating multiple sensors (cameras) to construct and optimize new synthetic indices; (v) integrate a trained deep learning network to implement a classification task. These examples demonstrate the tool’s ability to automate analysis, integrate heterogeneous data and models, and support reproducible computation of agronomically relevant traits. drone2report streamlines orthorectified UAV-image processing for precision agriculture by linking orthomosaics to standardized, plot-level outputs. Its modular, configuration-driven design allows transparent workflows, easy customization, and integration of multiple sensors within a unified analytical framework. By facilitating reproducible, multi-modal image analysis, drone2report lowers technical barriers to UAV-based phenotyping and opens the way to robust, data-driven crop monitoring and breeding applications. Full article

Accurately identifying medium- and low-yield farmlands (MLYF) and diagnosing their constraints are essential for targeted improvement of productivity and national food security. However, traditional evaluation is usually limited by coarse spatial resolution and high labor costs, and a methodological gap remains between large-scale MLYF classification and system constraints diagnosis. To address the current methodological gaps, this study developed a comprehensive framework to determine the spatial distribution of MLYF in northern China and clarify their key constraints. The framework combined the Spatio-Temporal Random Forest (STRF) algorithm with vegetation indices (VIs), climate, and soil data to delineate MLYF and uses interpretable machine learning to diagnose major constraints. The model showed high explanatory power and ensured the reliability of attribution results. The results showed that MLYF exhibited obvious spatial heterogeneity, accounting for 48.66% of the total cultivated land in the study area. These MLYF are primarily concentrated in the northwestern Loess Plateau (LP), the central Along the Great Wall (ATGW) region, and the peripheries of the Huang-Huai-Hai (HHH) Plain. In addition to spatial classification, our analysis revealed significant differences in constraint mechanisms: soil structural, nutrient, and salinization constraints predominantly restrict productivity in the HHH Plain, whereas water stress and soil erosion are the primary drivers of yield gaps in the LP and ATGW regions. These findings provide new data and insights for understanding the spatial heterogeneity of farmland quality in typical dryland agricultural regions in northern China, and offer a scientific basis for targeted land improvement and regional agricultural sustainability. Full article