Search Results (11,458)

Diffusion models have achieved remarkable success in image generation, but their iterative denoising process requires repeated evaluations of large neural networks, resulting in high inference latency. Recent training-free acceleration methods, such as DeepCache, exploit temporal redundancy in U-Net features by caching and reusing high-level representations across adjacent denoising steps. However, existing caching strategies usually adopt a static skip-branch selection throughout the sampling trajectory, ignoring the stage-dependent frequency evolution of diffusion sampling. In this paper, we propose an interval-guided adaptive branch routing strategy to improve training-free feature reuse. Motivated by the observation that low-frequency global structures change rapidly in early denoising stages while high-frequency details dominate later refinement, our method dynamically adjusts the skip branch according to the timestep. It preserves deeper computation in early stages for semantic reconstruction and progressively shifts to shallower branches in later stages to reduce redundant computation while maintaining fine-grained details. The proposed method requires no retraining and can be directly applied to pretrained U-Net-based diffusion models. Experiments show that FreqCache achieves up to 1.93× speedup on CIFAR-10, 1.50× speedup on LSUN-Bedroom and LSUN-Churches, and 10.60× speedup on ImageNet 256 × 256 compared with the baseline, while maintaining an Fréchet Inception Distance (FID) score comparable to or slightly better than DeepCache at the same cache interval. Full article

Background/Objective: Parental feeding practices play an important role in shaping children’s dietary intake, eating behaviours, and long-term health outcomes. Although several questionnaires assess feeding practices, few have been validated among socioeconomically disadvantaged populations, despite these groups being disproportionately affected by food insecurity and diet-related health inequities. This study assessed the structural validity and internal consistency of the Feeding Practices and Structure Questionnaire—Solid Feeding version (FPSQ-S)—among socioeconomically disadvantaged mothers and fathers of young children. Methods: Two cross-sectional online surveys were conducted with 178 mothers and 94 fathers of children aged 5–35 months living in disadvantaged households. Confirmatory factor analysis was used to examine the structural validity of the FPSQ-S. Internal consistency was assessed using Cronbach’s alpha and Hancock’s H coefficients. Results: The original six-factor FPSQ-S structure was retained and demonstrated acceptable overall model fit in this disadvantaged sample (CFI = 0.846, TLI = 0.821, RMSEA = 0.070). Internal consistency ranged from acceptable to excellent across subscales (Cronbach’s α = 0.63–0.93; Hancock’s H = 0.64–0.93). Most items loaded satisfactorily onto their intended constructs; however, two items within the Feeding on Demand construct demonstrated weak factor loadings, and this construct showed lower reliability than the remaining subscales Conclusions: This is the first study to evaluate the FPSQ-S among socioeconomically disadvantaged mothers and fathers of children aged 5–35 months. The FPSQ-S demonstrated acceptable structural validity and reliability. While the six-factor structure was largely supported, further refinement of the Feeding on Demand construct and additional psychometric evaluation are warranted. Full article

To improve the efficiency of TIG(Tungsten Inert Gas) welding, our team developed a novel fast-frequency pulsed twin-TIG welding power source and matched welding procedures to overcome the drawbacks of conventional high-efficiency TIG welding. After parameter optimization, stable, high-efficiency and high-quality welding of 316L stainless steel can be realized. Compared with traditional DC TIG welding, the mechanical properties of joints are greatly improved: the weld grain size is refined by 38% under moderate current, while tensile strength, elongation and microhardness rise by 13.6%, 26% and 10% respectively, which achieves simultaneous improvement in strength and ductility. Numerical simulations were carried out to analyze the evolution of molten pool temperature field and velocity vector flow field. The simulation results are highly consistent with experimental data, which verifies the reliability of the model and lays a foundation for the study of molten pool behavior. Combined with molten pool flow characteristics and weld microstructure, the evolution mechanism of microstructure and texture as well as grain refinement in this welding process is revealed. Full article

Accurate tropical cyclone (TC) track prediction remains challenging, as numerical models suffer from high computational cost, substantial storage requirements, and physical parameterization uncertainties, while data-driven large AI models depend heavily on training data volume and high-resolution inputs, resulting in prohibitive computational overhead. To address these issues, this paper proposes TCN-GAN-DM, a three-stage deep learning framework based on the China Meteorological Administration (CMA) Tropical Cyclone Best Track Dataset. Specifically, a dual-stream temporal convolutional network (TCN) first extracts temporal features from track and meteorological sequences, respectively. A generative adversarial network (GAN) then takes these features and produces multiple physically plausible candidate tracks via noise injection. Finally, a conditional diffusion model (DM) refines the predicted positions through progressive denoising. Experimental results for TCs in 2024 show that under the fair deterministic comparison using a single fixed candidate, the model achieves a 6 h track error of 49.10 km, which is comparable to CMA-GFS (49.75 km) and HWRF (44.34 km), and substantially lower than the large AI model FuXi (120.44 km). When evaluating the oracle metric (best-of-K, K = 6) as an upper bound of coverage, the model achieves the smallest errors among all models at 6 h (24.04 km) and 12 h (55.81 km). In addition, the proposed model has advantages over CMA-GFS, HWRF, and FuXi in terms of computational resource consumption and hardware deployment cost. However, its mean track error increases more rapidly beyond 12 h, and at lead times of 18 h and 24 h the model is outperformed by HWRF, FuXi, and CMA-GFS, indicating that its current strength lies primarily in short-term prediction. Consequently, the practical utility of TCN-GAN-DM is currently demonstrated for 6–12 h TC track prediction, offering a new solution for disaster prevention and mitigation that balances accuracy and deployment cost at these specific time scales. Full article

Joint detection of strawberry growth stages and defective fruit is needed for harvest planning and quality screening, but field images make this task difficult because stage-related visual differences are subtle, flowers and early fruits are often small and densely distributed, and occlusion weakens localization reliability. This study develops Multi-Scale Refined Detection and Classification YOLO (MRDC-YOLO), a lightweight detector based on the YOLO11s framework, for this fine-grained detection scenario. The backbone, neck, and detection head are redesigned with three modules: a Multi-Scale Adaptive Edge Enhancement Module (MAEM), a Reparameterized Progressive Feature Aggregation (RPFA) module, and a Decoupled Cross-Scan Head (DCSH). MAEM strengthens boundary and texture responses for visually similar categories, RPFA reduces redundant multi-scale fusion while maintaining features for dense small targets, and DCSH introduces task-aware classification and regression branches with cross-scan-inspired spatial modeling for occlusion-sensitive localization. Experiments on a five-class strawberry dataset containing 5114 images show that MRDC-YOLO achieves 95.63% mAP@0.5 and 82.39% mAP@0.5:0.95. Over YOLO11s, the model yields a 2.06-percentage-point gain in precision and 1.34- and 1.53-percentage-point gains in mAP@0.5 and mAP@0.5:0.95, together with 10.7% fewer parameters and 8.9% lower GFLOPs. These results suggest that MRDC-YOLO improves fine-grained category discrimination and localization while retaining a smaller model size than the YOLO11s baseline. Full article

Clarifying the sources of disparity and the systemic structure of influencing factors behind China’s economic resilience is crucial for promoting regional coordinated development and ensuring national security. This study constructs an evaluation index system based on the DPSIR model and employs the entropy method to measure China’s economic resilience from 2008 to 2023, examining its temporal evolution and spatial distribution. A bi-dimensional decomposition method of Gini coefficient is applied to examine disparities from both spatial and structural perspectives. Furthermore, the DEMATEL-ISM model is employed to reveal the systemic structure of influencing factors. The findings reveal that: (1) China’s economic resilience steadily improved during the study period, showing a spatial gradient of “Eastern > Central > Northeastern > Western,” with its geographic center shifting southeastward, reflecting strong spatial dependence. (2) Disparities in economic resilience have generally widened. Inter-regional differences are the main source of spatial disparities, while variations in response dominate the structural disparities. Initially, disparities were mainly due to differences in influence between eastern and western regions, but by the end of the period, disparities in driving forces became the key contributor. (3) Influencing factors follow a four-level, three-stage hierarchical structure. Foreign capital withdrawal risks, innovation investment, technological progress, factor supply, and the output of opening-up constitute deep-level factors influencing economic resilience. This study refines the evaluation framework of economic resilience and provides important references for understanding the disparities in China’s economic resilience and developing targeted improvement strategies. Full article

For hydrological parameterization in high-Andean catchments, it is necessary to understand whether near-surface hydro-structural soil properties can provide a surrogate signal of particle-size composition when direct texture information is sparse. This study evaluated the extent to which sand, silt, and clay fractions can be approximated from organic matter ($\mathrm{OM}$), bulk density (${\rho }_b$), and saturated hydraulic conductivity ($K_{\mathrm{sat}}$) in the Zamora Huayco (ZH) and Irquis catchments, southern Ecuador. A harmonized dataset ($n=44$) was analyzed through exploratory statistics, compositional assessment, correlation analysis, PCA, fraction-wise regression, $\mathrm{ILR}$-based modeling, AIC/BIC term reduction, sensitivity analysis excluding $\mathrm{OM}$, nested $\mathrm{LOOCV}$, and bootstrap-based uncertainty intervals. Among LULC classes, samples classified as paramo occupied a distinct high-Andean hydro-edaphic domain, characterized by a differentiated relationship between soil physical properties and hydrological behavior. PCA showed that the dominant covariance structure involved $\mathrm{OM}$, ${\rho }_b$, $K_{\mathrm{sat}}$, and the redistribution between sand and silt. The BIC-reduced $\mathrm{ILR}$ model provided the most balanced formulation, with positive nested $\mathrm{LOOCV}$ performance for sand, silt, and clay ($R_{\mathrm{LOOCV}}^2=0.147$, $0.704$, and $0.124$, respectively) and exact $100\%$ compositional closure after inverse transformation. Silt was the most stable predicted fraction, whereas sand and clay retained larger residual uncertainty, stronger tail departures, and partial compression of the observed variability. The proposed equations provide local hydro-pedotransfer support, although their predictive signal remains dependent on further refinement, uncertainty assessment, and external validation before regional application. Full article

Background: Gestational diabetes mellitus (GDM) affects between 1% and 14% of pregnancies worldwide. Major risk factors include advanced maternal age, excess adiposity, family history of type 2 diabetes, and unhealthy dietary habits. In Mexico, evidence on the association between dietary patterns and GDM remains scarce, particularly in socioeconomically vulnerable populations with limited access to specialized nutrition services. This study aimed to evaluate the association between dietary patterns and the presence of GDM in pregnant women attending the outpatient obstetrics clinic of a teaching public hospital in Guadalajara, México. Methods: We conducted a case–control study including 169 pregnant women: 71 with GDM confirmed by the ADA one-step 75 g oral glucose tolerance test OGTT criteria and 98 without GDM based on a negative OGTT, recruited consecutively from the same clinic during the same period. Dietary intake was assessed using a culturally adapted and validated Food Frequency Questionnaire. Dietary patterns were identified through Principal Component Analysis, and associations were examined using logistic regression adjusted for maternal age, pregestational BMI, and family history of type 2 diabetes. Results: Women with GDM had higher maternal age, greater pregestational BMI, and more frequent family history of type 2 diabetes compared with controls. Three dietary patterns were identified: Western, Healthy, and Dairy/Refined. High adherence to the Western pattern was inversely associated with GDM (aOR = 0.36; 95% CI: 0.16–0.78; p = 0.010); however, this finding most likely reflects post-diagnosis dietary modifications rather than a protective effect, while maternal age remained the strongest risk factor (OR = 1.09; 95% CI: 1.03–1.16; p = 0.002). The Healthy pattern (aOR = 1.25; 95% CI: 0.55–2.82; p = 0.593) and the Dairy/Refined pattern (aOR = 0.80; 95% CI: 0.39–1.66; p = 0.554) were not significantly associated with GDM in the adjusted model. Conclusions: GDM was associated with older maternal age, higher pregestational BMI, and family history of T2DM. The inverse association with the Western pattern may reflect post-diagnosis dietary changes rather than a protective effect. Due to the retrospective design, causal inference is not possible, highlighting the need for longitudinal studies. Full article

Background/Objectives: Preoperative differentiation among benign thyroid nodules, follicular variant papillary thyroid carcinoma (FV-PTC), and classical papillary thyroid carcinoma (C-PTC) remains clinically challenging. FV-PTC is particularly difficult to identify due to its substantial sonographic and cytological overlap with both benign nodules and other malignant subtypes, frequently resulting in overtreatment or delayed diagnosis. This study aimed to develop and validate an interpretable multimodal model for accurate three-class discrimination using routine ultrasound images, with a specific focus on improving the preoperative identification of FV-PTC. Methods: This retrospective study included 479 pathologically confirmed thyroid nodules from 462 patients. Conventional ultrasound features and radiomics features extracted from grayscale ultrasound and color Doppler flow imaging were used to construct three predictive models: a Conventional Ultrasound model (conventional ultrasound features only), a Radiomics model (radiomics features only), and an Integrated model (combined features). Each model was trained using four machine learning classifiers. Model performance was evaluated using the area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and F1 score. Model interpretability was assessed using SHapley Additive exPlanations (SHAP) analysis, and clinical usefulness was evaluated using decision curve analysis (DCA). Results: The support vector machine (SVM)-based Integrated Model achieved the best overall performance. In the independent testing cohort, the AUCs were 0.853 for FV-PTC, 0.882 for C-PTC and 0.928 for benign nodules. The Integrated Model showed the greatest improvement for FV-PTC, with a ΔAUC of 0.141 compared with the Conventional Ultrasound Model. SHAP (SHapley Additive exPlanations) analysis identified wavelet-HL_gldm_Dependence and wavelet-HH_glcm_InverseVariance as the two most important radiomics predictors in both the Radiomics Model and the Integrated Model, demonstrating robust cross-model stability and high discriminative power. Conclusions: The SVM-based Integrated Model demonstrated promising performance for three-class classification of thyroid nodules and enhanced the preoperative identification of FV-PTC. This approach may provide an interpretable and noninvasive decision-support tool for refining subtype-specific risk stratification and supporting individualized clinical management. Full article

Electron configurations are known to provide valuable insights into the electronic structure and behavior of atoms. They specify which and how the electronic (sub-) shells are occupied, and are thus an essential ingredient for most atomic observables. When combined with the shell model and the successive filling of shells, these configurations help explain the Periodic Table and much of chemical binding. They also establish a qualitative framework for analyzing excitation, ionization and relaxation processes and may facilitate a wide range of astrophysical and plasma simulations. Here, we review the role of electron configurations for understanding atomic behavior in interactions with particles and radiation. In particular, we identify several central requirements for an efficient treatment of configuration lists and define a domain-specific language in order to generate, manipulate and analyze such lists as well as to extract physically relevant information. We also demonstrate the implementation of this language in Jac, the Jena Atomic Calculator. An efficient handling of configurations will refine the coupling of structure codes with the spectral synthesis of plasma radiation, the setup of ionic cascades or even non-LTE plasma simulations. This common framework for dealing with electron configurations therefore improves consistency, reproducibility and scalability of atomic modeling. Full article

Nutrient stoichiometry, particularly the balance of carbon (C), nitrogen (N), and phosphorus (P), plays a fundamental role in regulating freshwater ecosystem dynamics, primary production, and biogeochemical cycling. This study presents one of the first dedicated reviews to combine bibliometric mapping with ecological synthesis of C:N:P ratios in inland waters, drawing on 1004 publications indexed in the Web of Science Core Collection (2000–2025), comprising peer-reviewed articles and review articles refined by document type, language, and research area. Bibliometric mapping using VOSviewer (version 1.6.20) identified exponential growth in publications after 2010, with phosphorus dynamics and eutrophication emerging as the most-cited themes, while recent years have shown increasing attention to C:P ratios as reliable ecological indicators. Four dominant thematic clusters were identified: Nutrient Cycling and Biogeochemistry; Phytoplankton and Food Web Dynamics; Eutrophication and Water Quality; and Climate Change and Ecosystem Responses. Ecological synthesis demonstrated substantial deviations from the canonical Redfield ratio (106C:16N:1P), with pronounced stoichiometric variability across trophic states, latitudes, and ecosystem types. Case comparisons revealed high C:P ratios in Arctic and alpine lakes linked to dissolved organic carbon inputs, low N:P ratios in tropical waters that promote cyanobacterial dominance, and stable, low phosphorus concentrations in deep African lakes. These findings emphasize the significance of flexible stoichiometry in predicting ecosystem tipping points, managing harmful algal blooms (HABs), and guiding nutrient restoration strategies. By integrating bibliometric and ecological evidence, this study identifies C:P ratios as a promising candidate indicator that merits further field validation for freshwater management, while underscoring persistent research gaps in microbial stoichiometry, cross-scalar modeling, and policy uptake in the Global South. Full article

Buds are continuously renewed sensory organs in which development, adult maintenance, and repair share overlapping molecular circuitry. During embryogenesis, WNT/β-catenin signaling promotes taste placode formation and placodal Shh expression, while SHH refines papilla spacing and restricts neighboring papilla formation. SOX2 functions as a taste-competence and progenitor maintenance factor. In adults, LGR5/LGR6–RSPO–WNT signaling sustains progenitor activity, and gustatory neurons are an important source of RSPO2; available genetic evidence is consistent with a neuron-derived contribution to the LGR5/LGR6 niche, and AAV-Cre-mediated neuron-specific ablation of Rspo2 in the petrosal ganglion led to near-complete loss of circumvallate taste buds. HH signaling from epithelial and neuronal sources further supports SOX2-dependent progenitor homeostasis. Lineage allocation is governed by transcriptional programs that include POU2F3/SKN-1a for sweet, umami, and bitter type II taste receptor cells, and ASCL1 with posterior-field NKX2-2 for type III presynaptic/sour cells. After denervation or irradiation, regeneration depends primarily on LGR5⁺/KRT14⁺ progenitors and may be supplemented, in specific injury contexts, by plasticity of a subset of K8-lineage taste receptor cells that acquire KRT14/SOX2/PCNA progenitor-like features. Key unresolved questions include the direct chromatin targets of taste lineage regulators (which remain to be defined by ChIP-seq in native taste progenitors), the identity of the type I cell selector, the contribution of dedifferentiation across injury models, and the degree to which mouse-derived networks are conserved in human taste biology. Full article

Microplastics are ubiquitous environmental particles with complex physical and chemical properties that enable them to interact with other contaminants. Recent evidence suggests that microplastics act as carriers for various chemical pollutants, altering their transport, deposition, and deposition dose. This conceptual review synthesizes current knowledge of radon progeny behavior and microplastic properties and suggests potential mechanisms for their interaction, although direct experimental validation of radon progeny specifically is currently lacking. It discusses attachment kinetics, transport pathways in air and water, and microplastic-mediated shifts in human lung deposition patterns and ecological exposure. Theoretical dosimetry reasoning suggests that, if attachment occurs, small respirable microplastics (1–10 μm) could increase inhalation doses by prolonging the airborne residence time of progeny indoors, whereas macro- and coarse microplastics would primarily affect localized environmental hotspots. These possibilities remain to be tested experimentally. Integrated experimental and modelling approaches, including radon chamber studies, aerosol and aquatic transport experiments, respiratory tract modelling, and ecological bioassays, are proposed to quantify these processes and inform risk assessment. Knowledge gaps remain in attachment efficiency, retention, co-contaminant interactions, and long-term exposure scenarios. Addressing these gaps is critical for refining human and ecological risk assessments and guiding regulatory frameworks in radon-microplastic-impacted environments. Full article

In the context of rural revitalization and carbon neutrality, this study addresses energy inefficiency and thermal discomfort in existing rural housing by optimizing passive design strategies for the “SunnyInside” sunroom model. Using parametric simulation with Ladybug and Honeybee, a dynamic light-thermal coupling model was developed to evaluate climate-adaptive performance in two distinct Chinese climates: the cold climate of Datong and the hot-summer-cold-winter climate of Wuhan. Multi-objective optimization focused on orientation, overhang depth, and photovoltaic (PV) tilt angles to enhance ventilation, shading, and daylighting. Key findings include: (1) Optimal building orientations of 15° west of south (Datong) and 16° east of south (Wuhan); (2) A 1.5m overhang depth in Wuhan improved summer shading efficiency by 28.6% and extended thermal comfort duration by 15%; (3) PV tilt ranges of 29–36° (Datong) and 13–23° (Wuhan) maximized energy performance. These optimizations achieved a 19.3–24.7% improvement in comprehensive performance coefficients and reduced air conditioning energy consumption by 17.8–21.4 kWh/m² (with ≥82% photovoltaic conversion efficiency). The study demonstrates the effectiveness of parametric simulation and intelligent algorithms in refining climate-responsive rural housing renovations, providing quantitative guidelines for PV shading systems across diverse climatic zones. Full article

Huanglongbing, a destructive citrus disease of global importance that is also present in Saudi Arabia, is associated with Candidatus Liberibacter asiaticus (CLas) and remains a major threat to citrus production. Although previous studies have documented sequence variation and prophage polymorphism in CLas, broader comparisons of prophage-associated gene content remain limited. In particular, comparative orthogroup analysis of prophage gene-content conservation across geographically structured CLas populations has rarely been explored. In this study, we analyzed 42 CLas prophage genomes from Saudi Arabia and other geographic regions using a comparative orthogroup framework. OrthoFinder assigned 99.1% of predicted proteins (1825 of 1841) to 64 orthogroups, with only 16 genes remaining unassigned. A small number of rare orthogroups restricted to only a few genomes were identified, and no orthogroup was detected in all genomes. Presence–absence analyses supported a predominantly conserved prophage gene repertoire together with a small accessory component, while also indicating that apparent absences should be interpreted in light of mixed assembly status and prophage-region completeness. Saudi Arabian genomes were distributed within the broader global framework and exhibited generally similar gene-content profiles rather than a deeply separated lineage. Functional interpretation of representative orthogroups identified conserved prophage-associated genes related to replication, helicase activity, and phage packaging, whereas variable orthogroups were primarily associated with hypothetical or accessory prophage-related functions. Overall, these results are consistent with a model in which CLas prophage diversification is associated more with sequence-level variation and localized structural differences than with extensive gain or loss of prophage genes. These findings further refine current understanding of CLas genome evolution and highlight conserved prophage-associated targets that may support molecular diagnostics and epidemiological surveillance. Full article