Search Results (2,824)

Existing appearance-based and video-based gaze estimation methods mainly rely on frame-wise prediction or local-window temporal fusion, which limits their ability to model long-range dependencies and to explicitly suppress output-level jitter. This leaves a gap in unified temporal gaze estimation frameworks that jointly address contextual feature aggregation and prediction-level stabilization. To address this limitation, we propose a unified state-space temporal gaze estimation framework to improve both angular accuracy and temporal consistency. Specifically, consecutive eye image sequences are mapped into a shared latent state space, where spatial appearance cues and inter-frame dynamics are jointly modeled. A feature-level temporal aggregation module is further designed to adaptively reweight historical observations for the current estimate, and a prediction-level temporal correction module is introduced to suppress short-term fluctuations while preserving rapid gaze shifts. On the TEyeD dataset after quality screening, the proposed method achieves a 3D gaze MAE of $0.533°$, compared with $0.96°$ for Model-aware and $3.18°$–$3.47°$ for the ResNet baselines reported in the original TEyeD paper, while maintaining manageable deployment overhead. These results indicate that the proposed framework provides a favorable balance between estimation accuracy, temporal stability, and practical efficiency. Full article

Accurate measurement of low-level radon concentrations in the environment is increasingly important for climate research, radon priority area delineation, and atmospheric studies. Adsorbent-based radon detectors offer high sensitivity but suffer from strong temperature dependence of radon adsorption and rapid degradation under humid conditions, limiting their applicability in long-term environmental monitoring. This work presents a universal design methodology for temperature- and moisture-compensated radon detectors based on hermetically packaged adsorbents enclosed by radon-permeable polymer foils. Analytical models describing the opposing temperature dependences of radon adsorption in adsorbents and radon permeability in polymers are combined to derive a general optimization criterion that minimizes temperature-induced response variations over a defined temperature range. The method is applicable to arbitrary combinations of adsorbent materials and polymer foils, provided their radon adsorption and permeability characteristics are known. The approach is demonstrated using activated carbon fabrics and common polymers (LDPE, HDPE, and polypropylene), for which optimal design parameters are identified. In addition, water vapor permeation through polymer foils is modeled to estimate moisture protection and permissible exposure durations under high humidity. The results demonstrate that appropriately designed compensation modules can significantly reduce temperature sensitivity while extending operational stability in humid environments, enabling next-generation high-sensitivity radon detectors suitable for environmental applications. Full article

Background/Objectives: Tirzepatide (TZP), a dual glucose-dependent insulinotropic polypeptide and glucagon-like peptide-1 receptor agonist, induces substantial weight loss in patients with obesity; however, pharmacologically induced weight reduction may be accompanied by losses in fat-free mass (FFM), muscle strength (MS), and resting metabolic rate (RMR), potentially influencing metabolic health. The metabolic implications of short-term preservation of metabolically active tissue during TZP therapy remain incompletely characterized. Methods: We performed a secondary, exploratory analysis of a previously published 12-week prospective, non-randomized comparative study including 60 patients with obesity treated with TZP (n = 30 TZP+Low Energy Ketogenic Therapy [LEKT]; n = 30 TZP+Low Calorie Diet [LCD]). Body weight (BW), fat mass (FM), FFM, MS, and RMR were assessed at baseline and week 12. Glycaemic parameters included fasting glucose, insulin, hemoglobin A1c (HbA1c), and HOMA-IR. All analyses were exploratory and hypothesis-generating. Results: Both groups achieved comparable reductions in BW after 12 weeks. FM decreased in both groups, while relative preservation of FFM, MS, and RMR was observed in one dietary context. Short-term changes in HbA1c, insulin, and HOMA-IR were statistically associated with concurrent changes in FFM, MS, and RMR, whereas no consistent associations were observed with changes in total BW or FM. Baseline glycaemic values were largely within the normoglycemic range. Conclusions: In this short-term secondary exploratory analysis, preservation of metabolically active tissue during TZP therapy was associated with concurrent glycaemic profiles, whereas no consistent associations were observed with total weight loss magnitude. These findings do not imply causality and should be interpreted as hypothesis-generating, warranting confirmation in larger, randomized, long-term studies. Full article

Adenosine triphosphate (ATP) serves as a central energy currency and signaling molecule in cellular processes, with ATP-binding sites in proteins playing critical roles in enzymatic catalysis, signal transduction, and gene regulation. The accurate identification of ATP-binding sites is essential for understanding protein function mechanisms and facilitating drug discovery, enzyme engineering, and disease pathway analysis. In this study, we present a novel hybrid deep learning framework that synergizes heterogeneous learning paradigms based on protein sequence information for accurate ATP-binding site prediction. Our approach integrates two complementary base classifiers. One is a Transformer-based model, which leverages high-level contextual embeddings generated by Evolutionary Scale Modeling 2 (ESM-2), a state-of-the-art protein language model, combined with a local–global dual-attention mechanism that enables the model to simultaneously characterize short-segment and long-range contextual dependencies across the entire protein sequence. The other is a deep Q-network (DQN)-inspired classifier that achieves residue-level prediction as a sequential decision-making process. The final predictions are generated using a weighted ensemble strategy, where optimal weights are determined via cross-validations to leverage the strengths of both models. The prediction results on benchmark independent testing sets indicate that our method achieves satisfactory performance on key metrics. Beyond predictive efficacy, this work uncovers the intrinsic biological mechanisms underlying protein–ATP interactions, including the synergistic roles of local structural motifs and global conformational constraints, as well as family-specific binding patterns, endowing the research with substantial biological significance. The research in this work offers a deeper understanding of the protein–ligand recognition mechanisms and supportive efforts on large-scale functional annotations that are critical for system biology and drug target discovery. Full article

Age estimation based on facial images is a challenging task due to the complex and nonlinear nature of facial aging, which is influenced by both genetic and environmental factors. To address this challenge, we propose a hybrid ConvNeXt–Transformer framework that combines convolutional local feature extraction with attention-based global contextual modeling within a unified age regression pipeline. The methodological contribution of this work lies in the sequential integration of these two complementary paradigms for facial age estimation, allowing the model to capture both fine-grained textural cues—such as wrinkles and skin spots—and long-range spatial dependencies. We evaluate the proposed framework on benchmark datasets including MORPH II, CACD, UTKFace, and AFAD. The results show competitive performance across these datasets and confirm the effectiveness of the proposed hybrid design through extensive ablation analyses. Experimental results demonstrate that our approach achieves state-of-the-art MAE on MORPH II (2.26), CACD (4.35), and AFAD (3.09) under the adopted benchmark settings while remaining competitive on UTKFace. To address computational efficiency, we employ ImageNet pre-trained backbones and explore different architectural configurations, including fusion strategies and varying depths of the Transformer module, as well as regularization techniques such as stochastic depth and label smoothing. Ablation studies confirm the contribution of each component, particularly the role of attention mechanisms, in enhancing the model’s sensitivity to age-relevant features. Overall, the proposed hybrid framework provides a robust and accurate solution for facial age estimation, effectively balancing performance and computational cost. Full article

Non-destructive and automated detection of bacterial contamination is a critical prerequisite for ensuring high efficiency production and quality control in plant tissue culture. In this study, we developed a multispectral image acquisition system for Alocasia explants and proposed a novel image fusion model, termed Intensity-Texture enhanced Swin Fusion (ITSF). The ITSF framework employs convolutional neural networks to extract texture and intensity features from visible and near-infrared channels. Subsequently, a Swin Transformer-based module is integrated to model long-range spatial dependencies, ensuring cross-domain integration between the texture and intensity features. We formulated a composite loss function to guide the fusion process toward optimal results. This objective function integrates texture loss, entropy weighted structural similarity index (SSIM) and intensity aware dynamic gain guided loss. Experimental results demonstrate that the proposed method significantly enhances the visual saliency of bacteria and achieves superior quantitative performance across a comprehensive range of objective image fusion metrics. The detection performance reached a mean Average Precision (mAP50) of 0.949 with the fused images, satisfying industrial requirements for high-precision inspection, which provides a critical technical solution for the industrialization of automated micropropagation. Full article

Remote sensing (RS) images are often degraded by atmospheric haze, which compromises both visual interpretation and downstream applications. To address this, we introduce CSGL-Former, a novel Cross-Stripes Global–Local Fusion Transformer for RS image dehazing. Our model efficiently captures anisotropic long-range dependencies using cross-stripes attention (CSA) and aggregates hierarchical global semantics via a Multi-Layer Global Aggregation (MLGA) module. In the decoder, global context is adaptively blended with fine-grained local features to restore intricate textures. Finally, inspired by the atmospheric scattering model, a soft reconstruction head restores the clear image by predicting spatially varying affine parameters, strictly preserving content fidelity while effectively removing haze. Trained end-to-end, CSGL-Former demonstrates a compelling balance of accuracy and efficiency. Extensive experiments on the RRSHID and SateHaze1K benchmarks show that our model achieves state-of-the-art or highly competitive performance against representative baselines. Ablation studies further validate the effectiveness of each proposed component. Full article

Accurate fault diagnosis of rotating machinery in complex industrial environments demands an optimal trade-off between feature representation capability and computational efficiency. Existing single-modality models relying solely on 1D time-series signals or heavy 2D time-frequency images often fail to simultaneously capture high-frequency transient impacts and long-range degradation trends. CLiST (Complementary Lightweight Spatiotemporal Network), a novel lightweight multimodal framework driven by time-frequency fusion, was proposed to overcome this limitation. The architecture of CLiST employs a synergistic dual-stream design: a LightTS module efficiently extracts global operational trends from 1D vibration signals with linear complexity, while a structurally pruned LiteSwin integrated with Triplet Attention captures local high-frequency textures from 2D continuous wavelet transform (CWT) images. This mechanism establishes explicit cross-dimensional dependencies, effectively eliminating feature blind spots without excessive computational overhead. The experimental results show that CLiST not only achieves perfect accuracy on the fundamental CWRU benchmark but also exhibits exceptional spatial generalization when independently evaluated on non-dominant sensor axes of the XJTUGearbox dataset. Furthermore, validation on the real-world dataset (Guangzhou port) proves that the framework has excellent robustness to the attenuation of the signal transmission path and reduces the performance fluctuation between remote measurement points. Ultimately, CLiST delivers highly reliable AI-driven image and signal-processing solutions for vibration monitoring in industrial equipment. Full article

Under dynamic loading conditions, the output voltage of proton exchange membrane fuel cells (PEMFCs) exhibits nonlinear degradation characterized by non-Gaussian fluctuations, abrupt changes, and long-range temporal dependence, which are difficult to model using conventional short-correlation or remaining useful life (RUL) prediction approaches. To capture both historical dependency and stochastic jump behavior, this study proposes a SHAP-driven mechanism–data fusion fractional stochastic degradation model based on fractional Brownian motion (fBm) and fractional Poisson process (fPp) for degradation trend forecasting. A terminal voltage mechanism model considering activation, ohmic, and concentration polarization losses is first established, and SHapley Additive exPlanations (SHAP) analysis is employed to quantify the contributions of multi-source operational variables and enhance interpretability. The Hurst exponent is then used to verify long-range dependence and jump characteristics in the voltage sequence. Subsequently, fBm is integrated with a fPp to construct a unified stochastic degradation framework capable of jointly describing continuous decay and discrete abrupt variations, enabling multi-step probabilistic prediction with confidence intervals. Validation on the publicly available FCLAB FC1 and FC2 datasets shows that the proposed model achieves superior overall performance under both steady and dynamic conditions, with MAPE/RMSE/R² of 0.027%/0.00178/0.9895 and 0.056%/0.00259/0.9896, respectively, outperforming fBm, Wiener, WTD-RS-LSTM, and CNN-LSTM methods. The proposed approach provides accurate and interpretable degradation forecasting for PEMFC health management and maintenance decision support. Full article

Background and Objectives: This study aimed to evaluate the effectiveness of Transfer of Energy Capacitive and Resistive (TECAR) therapy in treating active myofascial trigger points (MTrPs) in the upper trapezius muscle (UT) and to compare it with the effects of dry needling (DN). Materials and Methods: We recruited 29 men (mean age: 35.52 ± 5.73 years) with active MTrPs in the UT. Participants were randomly assigned to two groups: TECAR (n = 17) and DN (n = 12). Treatment was administered twice, with a 7-day interval between sessions. PPT, pain intensity (NRS), UT muscle strength (dynamometer), and cervical spine range of motion (ROM) were measured before treatment, immediately after each therapy session, and at a 30-day follow-up. Data were analyzed using parametric or non-parametric tests depending on data distribution (p < 0.05). Results: Both groups showed significant increases in PPT, but TECAR reduced NRS significantly more than DN (p < 0.001), demonstrating superior immediate analgesia. While TECAR temporarily decreased unaffected UT strength, it provided broader improvements in cervical mobility (flexion: 19.5%, contralateral rotation: 13.1%). Over 30 days, both groups improved PPT (TECAR: ~110%; DN: ~63%) and NRS (TECAR: ~97.1%; DN: ~84.5%). The TECAR group consistently outperformed DN in long-term pain reduction and achieved more substantial improvements in ROM. Conclusions: TECAR therapy appears to provide immediate and longer-term analgesic effects in the treatment of active MTrPs in the UT, although its impact on cervical ROM seems relatively limited compared with DN. It may therefore represent a useful, though less commonly applied, option for MTrPs management. Full article

Long-fiber thermoplastic (LFT) materials are a versatile category of composite materials that can be directly compounded (LFT-D) in twin screw extruders and compression molded. Originating in the automotive sector, the LFT-D process is becoming increasingly attractive for other industries where low cycle times, lightweight performance and recyclability are required. The purpose of this work is to summarize mechanical properties and findings from the investigations into LFT-D process–microstructure–property relationships and present a design of experiments (DoE) study based on the current state of the art. Primary parameters from LFT-D compounding, screw speed, fiber roving amount and polymer throughput m_p are chosen as DoE factors. Polyamide 6 (PA6) is reinforced with a glass fiber (GF) mass fraction w_f between w_f = 20% and w_f = 60%. Tensile, flexural and impact properties are chosen as DoE output parameters, characterized and discussed in relation to the state of the art. The unique microstructure of LFT-D materials, especially the existence of a charge and flow area as well as the fiber migration, is considered in the discussion. All mechanical properties characterized have a linear relation to w_f. This study demonstrates the interactive relationship between the main factors and w_f, which significantly influences the mechanical properties. This dependence of w_f on the DoE factors is accounted for in advanced response contour plots proposed in this work. Parameter recommendations for the screw speed are reported by ranges of w_f and polymer throughput for the goal of maximum mechanical properties or low coefficient of variations. At w_f < 30% a low screw speed is recommended to improve most mechanical properties as well as the coefficient of variation. Full article

Classroom humor is an important instructional strategy that enhances teaching effectiveness and improves student engagement. However, its automatic detection remains challenging due to the strong contextual dependency and implicit semantic shifts that characterize humorous expressions in teaching discourse. Conventional pretrained language models capture global semantics but often fail to focus on the subtle humor anchors that trigger incongruity. To address this issue, we propose T-HumorAGSA, a cognitive-inspired classroom teacher humor language detection model. The model employs BERT for contextualized semantic encoding, followed by a Gated Anchor-Guided Self-Attention (AGSA) mechanism that adaptively amplifies anchor-related features responsible for humor generation. A bidirectional gated recurrent unit (BiGRU) layer is further integrated to model long-range temporal dependencies within teaching utterances. T-HumorAGSA is evaluated on five datasets, including SemEval 2021 Task 7-1a, ColBERT, CCL2018, CCL2019 and the self-constructed teacher humor language dataset (T-Humor), demonstrating consistently strong performance. For instance, it achieves 0.9874 F1 on ColBERT and 0.9508 F1 on SemEval 2021 Task 7-1a, both outperforming the best baseline models. On the T-Humor dataset, the model attains a high F1 score of 0.9895, validating its capacity to detect subtle humorous cues in instructional discourse. The results demonstrate that the proposed model delivers excellent performance in classroom humor detection. Full article

Object detection in UAV-based fire rescue scenarios faces multiple challenges, including densely distributed small targets, severe occlusion, and interference from smoke and flames. Existing mainstream detection models, such as the YOLO series, often prioritize inference speed at the expense of modeling global context and spatial positional information, resulting in limited performance in such complex environments. To address these limitations, this paper proposes FirePM-YOLO, an object detection architecture optimized for fire rescue applications. Based on the YOLO framework, the proposed model introduces two key innovations: first, a Position-Aware Enhanced Mamba module (PEMamba) is designed, which incorporates a compact positional encoding mechanism, lightweight spatial enhancement, and an adaptive feature fusion strategy to significantly improve scene perception while maintaining computational efficiency. Second, a PEMBottleneck structure is constructed, which dynamically balances local convolutional features and global PEMamba features via learnable weights. This module is embedded into the shallow layers of the backbone network, forming an enhanced PEM-C3K2 module that captures long-range dependencies with linear complexity while preserving fine local details, thereby enabling holistic contextual understanding of fireground environments. Experimental results on the self-built “FireRescue” dataset demonstrate that compared with the original YOLOv12 and other mainstream detectors, the proposed model achieves improvements in both mean average precision (mAP) and recall while maintaining real-time inference capability. Notably, it exhibits superior detection performance on challenging samples, such as small-scale and partially occluded professional firefighting vehicles. Full article

Suicide is a major public health problem worldwide, and its reported reasons and methods show marked variation by gender and age. Although suicide rates are generally higher among men, suicides among women demonstrate distinct sociodemographic and age-related patterns that remain insufficiently explored. In Türkiye, national suicide statistics are available; however, nationwide, age-stratified analyses focusing exclusively on women are limited. This study aimed to investigate long-term trends, age-related differences in reported reasons and methods of suicide among women in Türkiye, and to provide insights relevant to age- and gender-sensitive prevention strategies. This retrospective, nationwide descriptive study analysed female suicide data obtained from the Turkish Statistical Institute between 2009 and 2023. A total of 12,868 female suicide cases were included (mean age 36.5 ± 19.3 years). Data were evaluated according to year, age group, marital status, educational level, suicide cause, and suicide method. Causes and methods were classified based on official administrative categories. Descriptive statistics were calculated, and associations between age groups and suicide causes and methods were assessed using Pearson’s chi-square test. During the 15-year study period, 12,868 women died by suicide in Türkiye. The annual suicide rate ranged from 1.81 to 2.46 per 100,000 population, with the lowest rate observed in 2017 and the highest in 2022. Among all age groups, the most frequent cause of suicide was illness, especially in women aged 45 and older. The proportion of suicides due to illness was 13.9% in the 15–24 age group, 24.6% in 25–34, 41.0% in 45–54, and 42.3% in 55–64 (p < 0.001). Emotional and relationship-related causes were more prevalent among younger women, particularly in the 15–24 age group (4.8%), but declined significantly with age (p < 0.001). Economic hardship was the least cited cause overall, especially among women under 35 (p < 0.001). Regarding methods of suicide, hanging was the most common method in all age groups and increased with age—35.8% in 15–24, 55.1% in 45–54, and 63.5% in 75+ age group (p < 0.001). The use of chemical substances peaked in the 15–24 age group (12.4%) and declined in older women (5.8% in 75+). Firearm use showed a significant inverse relationship with age, from 24.6% in those under 15 to 0.8% in women aged 75 and over (p < 0.001). These age-related differences in both the causes and methods of suicide were statistically significant (p < 0.001). Female suicide in Türkiye exhibits pronounced age-dependent differences in both causes and methods. Illness-related suicides and hanging predominate in older age groups, while younger women show a more diverse pattern of reported reasons and methods. The high prevalence of nonspecific classifications highlights limitations in current suicide reporting systems. These findings underscore the need for improved suicide classification, enhanced surveillance, and age- and gender-sensitive prevention strategies tailored to women across the lifespan. Full article

Dynamic Spectrum Access (DSA) has emerged as a viable solution to address spectrum scarcity in shared-spectrum networks. In response, the FCC established the Citizens Broadband Radio Service (CBRS) to manage and facilitate shared use of the federal and non-federal spectrum in a three-tiered access and authorization framework. However, due to the open nature of spectrum access and the usually limited coverage of the monitoring infrastructure, enforcing access rights in a shared-spectrum network becomes a daunting challenge. In this paper, we stipulate the use of crowdsourcing as a viable approach to engaging volunteers in spectrum monitoring in order to enforce spectrum access rights robustly and reliably. The success of this approach, however, hinges strongly on ensuring that spectrum access enforcement is carried out by reliable and trustworthy volunteers within the monitored area. To this end, a hybrid spectrum monitoring framework is proposed, which relies on opportunistically recruiting volunteers to augment the otherwise limited infrastructure of trusted devices. Although a volunteer’s participation has the potential to enhance monitoring significantly, their mobility may become problematic in ensuring reliable coverage of the monitored spectrum area. To ensure continued monitoring, inspite of volunteer mobility, deep learning-based models are used to predict the likelihood that a volunteer will be available within the monitoring area. Three models, namely LSTM, GRU, and Transformer, are explored to assess their feasibility and viability to predict a volunteer’s availability likelihood over an extended time interval, in a given spectrum monitoring area. Recurrent Neural Networks (RNNs) such as GRU and LSTM are effective for tasks involving sequential data, where both spatial and temporal patterns matter, which is the focus of volunteer availability prediction in spectrum monitoring. Transformers, on the other hand, excel at handling long range dependencies and contextual understanding. Furthermore, their parallel processing capabilities allows faster training and inference compared to RNN-based models like GRU and LSTM. A simulation-based study is developed to assess the performance of these models, and carry out a comparative analysis of their ability to predict volunteers’ availability to monitor the spectrum reliably. To this end, a real-world trace dataset of volunteers’ location, collected over five years, is used. The simulation results show that the three models achieve high prediction accuracy of volunteers’ availability, ranging from 0.82 to 0.92. The results also show that a GRU-based model outperforms LSTM and Transformer-based models, in terms of accuracy, Root Mean Square Error (RMSE), geodesic distance, and execution time. Full article