Search Results (5,461)

Source-free domain adaptation (SFDA) for cross-modality prostate MRI segmentation is challenging because source data are unavailable and pseudo-labels on target ADC images are often noisy. To address this problem, we propose Curriculum Anatomy-guided Learning Method with Population Template Priors (CALM), a source-free adaptation framework for this task. CALM constructs a population template prior from target predictions using top-k consensus aggregation and cross-round exponential moving average, then combines this prior with instance-level predictions through Soft-AND fusion. A high-confidence background constraint is further introduced to provide reliable negative supervision, and a coverage-driven curriculum is used to expand training from easy to hard cases based on pseudo-label/template agreement. This design forms an iterative process in which prior refinement and sample-reliability refinement reinforce each other during adaptation. Experiments on the PI-CAI dataset under the T2W-to-ADC setting show that CALM achieves an average Dice score of 73.63% and outperforms representative SFDA baselines in both segmentation accuracy and boundary quality. Ablation and model analyses support the contribution of each component. These results suggest that population-level anatomical priors can provide practical structural guidance for source-free cross-modality adaptation. Full article

Background: Artificial intelligence (AI) is transitioning from experimental pilots to core public health functions such as disease surveillance, resource planning, and analysis of social and structural determinants of health. Yet, health data collection and stewardship remain fragmented across the globe; some jurisdictions still rely on paper-based systems, while others operate noninteroperable digital systems that can exacerbate inequities. Treating health data as a global good therefore requires governance that enables innovation while protecting rights, safety, and trust. This study aims to develop a conceptual meso-level capability framework that translates responsible AI principles into organizational practices for public health agencies. Methods: We developed the framework using a targeted narrative synthesis of contemporary governance guidance and documented early implementation experiences, purposively selected to represent major strands of current practice and debate. A structured expert panel consultation (n = 9) was subsequently conducted to assess the face validity and content validity of the proposed framework domains. Results: We propose the Public Health Responsible AI Capability (PH-RAIC) framework, which adapts principles of transparency, accountability, fairness, ethics, and safety to institutional realities faced by public health agencies. PH-RAIC identifies four interdependent capability domains: (1) strategic governance and alignment; (2) data and infrastructure stewardship; (3) participatory design, equity, and public engagement; and (4) lifecycle oversight, learning, and decommissioning. All four domains achieved Content Validity Index (CVI) values ≥ 0.85 in the expert panel consultation. The framework is presented as a conceptual, meso-level model that has undergone preliminary expert validation but requires further empirical testing in real-world agency settings. Conclusions: PH-RAIC links these domains to example practices, diagnostic questions, and illustrative measurement indicators to help agencies navigate efficiency–equity trade-offs and strengthen legitimacy and accountability in AI-enabled public health systems. It offers a validated conceptual basis for future empirical testing and operational readiness tools. Full article

Background: Antibiotic overuse in hospitals is common and linked to adverse outcomes and antimicrobial resistance. Antimicrobial stewardship programs (ASP) aim to optimize prescribing and require context-specific adaptation. Objectives: To describe the experience of implementing and strengthening ASP in hospitals from six Latin American countries by using an integrated framework. Methods: The intervention included a point-prevalence survey (PPS) of antibiotic use, a baseline checklist, a continuous online education program, and individual facility meetings to share SWOT analyses and recommendations. The latter was performed based on PPS and checklist results. The checklist covers six domains (authorities’ commitment, organization, structure, and accountability; interventions; education and training; monitoring and surveillance; and internal communication). The training program spanned 12–18 months and addressed core ASP components. Results: The PPS across 67 hospitals showed an antibiotic use prevalence of 47.9%, with 63% of prescriptions deemed appropriate. The median checklist score was 61.2%. Among the categories assessed, monitoring and surveillance achieved the highest score (median 75.0; IQR 63.9–84.0), while education received the lowest (median 43.8; IQR 29.7–62.5). A total of 80 country groups and 35 individual hospital meetings were held. Conclusions: An integrated, data-driven framework combining PPS, checklists, individual hospital meetings, and sustained training provides a scalable approach to strengthening ASP in diverse Latin American hospitals, aligning with Pan American Health Organization (PAHO) guidance and global recommendations. Full article

Lunar optical and synthetic aperture radar (SAR) imagery provide complementary information for characterizing the lunar surface. However, their joint use remains challenging because of substantial cross-modality differences and severe illumination constraints, particularly in polar regions. To address this challenge, we propose LS2ODiff (Lunar SAR-to-Optical Diffusion), a diffusion-based framework designed for SAR-to-optical image translation in lunar environments. LS2ODiff uses SAR observations as conditional guidance in the diffusion process and incorporates a partial-convolution strategy into the U-Net backbone to handle irregular invalid regions. In addition, self-attention modules are incorporated into the downsampling stages of the U-Net to model long-range spatial dependencies and enhance global structural consistency in complex lunar terrain. We further construct a dedicated paired dataset of the lunar south polar region by registering Chandrayaan-II DFSAR data with Lunar Reconnaissance Orbiter (LRO) Narrow-Angle Camera (NAC) imagery. Comparative experiments against Pix2Pix, CycleGAN, SynDiff, and ConDiff demonstrate that LS2ODiff achieves better visual fidelity and quantitative performance in terms of peak signal-to-noise ratio (PSNR), structural similarity index measure (SSIM), Fréchet inception distance (FID), and learned perceptual image patch similarity (LPIPS). These results demonstrate the potential of diffusion models for high-fidelity lunar image translation, offering new opportunities for polar terrain interpretation and future exploration missions. Full article

Background/Objectives: Simulation-based education is increasingly used in health sciences to promote active learning and the development of clinical and non-technical skills. However, its implementation in undergraduate dental education remains limited. This study aimed to explore dental students’ perceptions of the Self-Learning Methodology in Simulated Environments (MAES©) applied to high-fidelity simulation. Methods: A mixed-methods, cross-sectional pilot study was conducted with 80 fourth-year dental students enrolled in a Pediatric Dentistry course at a Spanish university. Quantitative data were collected using a validated satisfaction questionnaire (Cronbach’s alpha = 0.905), and descriptive statistics were performed. Qualitative data were obtained through open-ended questions and analyzed using inductive content analysis. Results: Students reported high levels of satisfaction, motivation, and perceived learning, with mean scores above 8.5 out of 10 across all evaluated dimensions. The facilitator’s role received the highest ratings. Qualitative analysis identified four main themes: perceived advantages of the methodology, increased engagement and participation, the value of structured debriefing, and areas for improvement related to group dynamics and performance-related stress. Conclusions: The MAES© methodology was well received and perceived as a feasible approach in dental simulation-based education. It may support student-centered learning, collaboration, and reflective practice, providing practical guidance for educators interested in implementing active learning strategies. As an exploratory pilot study conducted in a single institution, these findings should be interpreted cautiously and warrant further research. Full article

Soil salinization is one of the most severe forms of land degradation in arid and semi-arid regions, posing substantial threats to agroecosystem stability and food security. In this study, saline–alkali soil collected from the Wuding River Basin in Yulin, Shaanxi Province was used to construct a three-factor amendment system comprising superabsorbent polymers (SAP), biochar, and humic acid. A systematic assessment was conducted to elucidate their combined effects on soil water–salt transport and crop growth. Results from one-dimensional constant-head infiltration experiments using indoor soil columns demonstrated that the application of amendments significantly increased cumulative infiltration and improved the uniformity of wetting-front advancement. Specifically, the treatments regulated the redistribution of salts within the soil profile; while surface salinity reduction varied, the leaching efficiency was significantly enhanced in the A2B2C2 treatment. Soil bulk density (BD) showed dynamic fluctuations during the growth cycle, peaking at 1.628 cm⁻³ during the branching stage, while high-rate biochar (A3) reduced BD by up to 13.64% compared to the control by the initial flowering stage. Fitting results based on the Philip and Kostiakov models further indicated that the combined amendment strategy—particularly the A2B2C2 treatment (30 kg/ha SAP, 15,000 kg/ha biochar, and 600 kg/ha humic acid)—markedly enhanced both the initial infiltration rate and the steady infiltration capacity. Field experiments corroborated the indoor findings: plant height and dry biomass of Melilotus officinalis (L.)Lam. were significantly higher under amendment treatments than in the control, driven by improved water availability, mitigated salt stress, and enhanced soil structure. Single-factor and multi-factor interaction analyses revealed that SAP exerted pronounced effects during early growth stages, whereas biochar and humic acid contributed more substantially during the middle to late stages through sustained regulatory functions. Collectively, the results demonstrate that the combined application of SAP, biochar, and humic acid improves the water–salt regime of saline–alkali soils through a coupled “water–salt–structure–plant” mechanism, ultimately enhancing crop productivity. This study provides both theoretical insights and practical guidance for the amelioration of saline–alkali soils. Full article

The composition and interface quality of jointed concrete can significantly influence chloride ion penetration, especially in coastal environments. This study investigates the transport behavior of chloride ions in concrete flexural members with varying joint configurations—no joint, smooth wet joint, and roughened wet joint—under different bending loads. After 28 days of curing, specimens were subjected to bending loads and immersed in an 8% NaCl solution for 300 days. Chloride ion concentrations were then measured at different depths and locations. Results revealed that joints, particularly smooth wet joints, significantly accelerate chloride ion transmission, and that chloride accumulation at the joint is consistently higher than in adjacent areas or jointless concrete. The apparent diffusion coefficient of chloride ions was notably higher at joint interfaces and increased with bending load level due to microcrack formation. Notably, the incorporation of Cementitious Capillary Crystalline Waterproofing (CCCW) in the concrete mix improved resistance to chloride ion penetration. A dosage of 1% CCCW proved most effective, reducing the diffusion coefficient at the joint by approximately 10%—demonstrating an optimal balance between performance and material efficiency. These findings provide practical guidance for improving the durability of jointed concrete structures in chloride-rich environments. Full article

Recent advances in medical image segmentation have introduced a wide spectrum of deep learning paradigms, including 2D/3D convolutional neural networks (CNNs), transformer-based architectures, vision-language models (VLMs), prompt-driven foundation models such as Segment Anything Model (SAM), and diffusion-based approaches. Although these methods have demonstrated remarkable performance across MRI, CT, PET, ultrasound, and endoscopic imaging, the rapid proliferation of architectures has created methodological uncertainty regarding optimal model selection under varying clinical and data constraints. Existing surveys primarily focus on architectural categorization, yet provide limited guidance for decision-oriented model selection. This study presents a comprehensive and decision-guided survey that systematically analyzes segmentation paradigms across imaging modalities, task types, dataset characteristics, and evaluation protocols. Beyond taxonomy, we propose a practical model selection framework that links clinical scenarios, such as small lesion detection, multi-organ 3D segmentation, limited-data regimes, and domain shift, to appropriate segmentation strategies. Furthermore, robustness, generalization, annotation variability, and benchmarking reproducibility are critically examined. By integrating architectural taxonomy, cross-modal comparative analysis, and a structured decision framework, this work provides a clinically oriented roadmap for selecting segmentation methods and highlights future research directions toward reliable and reproducible medical AI systems. Full article

This study experimentally investigates the evolution of natural frequencies of premium threaded connections under varying interface contact stiffness, aiming to establish a non-destructive vibration-based method for evaluating sealing contact conditions. The sealing interface features a sphere-on-cone configuration, and Hertzian contact theory is used to derive the contact pressure distribution, which shows a nonlinear increase in peak pressure with increasing normal load. Modal experiments were conducted under free–free boundary conditions using an impact hammer on a Φ88.9 mm × 6.45 mm P110 premium threaded connection. Three make-up torque levels (4081 N·m, 4393 N·m and 4691 N·m) were applied to create distinct contact states, and the first five orders of natural frequencies were extracted from the measured acceleration responses, using frequency response function (FRF) analysis with peak-picking identification. The results demonstrate that natural frequencies increase significantly with make-up torque, following a power-law relationship f = αT^β with R² > 0.97 for the first three modes. A critical torque range of 4200–4400 N·m is identified, below which frequencies rise sharply and above which the increase slows due to contact stiffness saturation. Lower-order modes are more sensitive to contact stiffness variations than higher-order modes. The findings confirm that natural frequency can serve as an effective non-destructive indicator for assessing tightening quality and detecting loosening in premium threaded connections, offering practical guidance for torque optimisation and structural health monitoring in oilfield operations. Although only three torque levels are used, the observed trend is physically consistent with contact mechanics theory and widely reported joint stiffening behavior. Therefore, the fitted relationship should be interpreted as a physically guided empirical model rather than a purely statistical fit. Full article

This study evaluates the adoption and efficacy of the 5th Dimension Building Information Modelling (5D-BIM) as a cost dimension for mega rail projects, extending the discussion beyond just technological implementation to consider broader policy and practical implications. The purpose of this article is to understand the governance context of 5D-BIM implementation for rail and transport projects and evaluate the effectiveness of the 5D-BIM framework as currently applied by conducting semi-structured interviews with key stakeholders. Drawing on semi-structured interviews with 22 stakeholders across government, industry, and technology providers, the research examines current 5D-BIM practices. While the primary focus of the research is 5D BIM implementations within the state of Victoria, Australia, which is currently experiencing a surge in rail projects, interviews were also conducted with additional stakeholders from international rail projects for context. The findings reveal fragmented adoption, varying levels of organisational maturity, and significant policy and implementation gaps, particularly in the role of government as the primary client of transport infrastructure. The results of the interviews emphasise the centrality of government and regulatory context in driving the adoption and implementation of 5D-BIM as the primary client of transportation infrastructure and identify actionable recommendations for policymakers and practitioners towards a more integrated approach to 5D-BIM in mega rail projects. While 5D-BIM demonstrates clear benefits in enhancing cost estimation, coordination, and decision-making, its effectiveness is constrained by the absence of clear standards, limited BIM literacy, and inconsistent regulatory guidance. This study provides one of the first empirical validations of the 5D-BIM governance framework, demonstrating that its success is driven less by technological capability and more by policy alignment, standardisation, and institutional leadership. Full article

The peak–end rule proposes that retrospective evaluations depend on the emotional peak and the end of an experience rather than on its duration. Two short, controlled vocabulary-learning experiments tested whether optimizing these moments improves retrospective emotional valence. Study 1 (N = 32) manipulated task length (4 vs. 8 words). Retrospective emotional valence did not differ significantly between groups (p = 0.459, d = 0.27), a result consistent with duration neglect under this short task–episode manipulation but not a strong test of pure temporal duration neglect. Retrospective emotional valence correlated more strongly with the peak–end mean than with the mean of reconstructed page-level ratings (r = 0.761 vs. r = 0.314; Steiger’s Z = 3.03, p = 0.002). Study 2 (N = 56) used a 2 × 2 design to optimize the candidate peak-related completion page and the structurally defined end check-in page through color and anthropomorphic graphics. Both peak (ηp² = 0.18) and end (ηp² = 0.22) optimization enhanced retrospective emotional valence, with a significant non-additive interaction (ηp² = 0.09): the effect of optimizing one node was reduced when the other node had already been optimized. For learning accuracy, the main effect of peak optimization was significant (F(1, 52) = 4.44, p = 0.040), but only the combined peak-and-end optimization significantly outperformed the control condition (p = 0.041, d = 1.11); neither single-optimization condition significantly differed from the control condition after correction. The findings provide preliminary evidence for a peak–end-consistent evaluation pattern in brief, controlled vocabulary-learning tasks, identify a non-additive interaction in peak–end optimization, and offer guidance for designing key interactive moments within similarly short, task-based learning episodes. Full article

Short-term precipitation forecasting is essential for disaster prevention, urban management, and weather-sensitive decision making, yet radar-based nowcasting remains challenging because precipitation systems evolve nonlinearly and high-frequency echo structures are easily over-smoothed by deterministic sequence models. This paper proposes a ViT-modulated diffusion spatiotemporal prediction network (VSTPN) that cascades a spatiotemporal prediction module with a ViT-conditioned diffusion refinement module. The spatiotemporal module models the temporal evolution of radar echoes, whereas the ViT-Diffusion module uses global contextual features as conditional guidance during iterative denoising to refine spatial structures. Experiments on the HKO-7 benchmark show that VSTPN achieves lower MSE and higher SSIM than the tested baselines and improves CSI, HSS, and POD at the evaluated reflectivity thresholds. At the 40 dBZ threshold, the model improves CSI, HSS, and POD, while its FAR is slightly higher than that of ETCJ-PredNet, indicating a recall–false alarm trade-off for intense echoes. Additional post-hoc diagnostic analyses of relative gains, metric consistency, threshold sensitivity, and component effect sizes further support the stability of the reported improvements under the current experimental protocol. The results suggest that coupling spatiotemporal sequence modeling with diffusion-based radar echo refinement is a feasible direction for short-term precipitation forecasting; nevertheless, probabilistic uncertainty evaluation, multi-domain validation, and additional generative-quality metrics remain important directions for future work. Full article

Remote sensing change detection (RSCD) is often disturbed by nuisance appearance variations, which can introduce pseudo-changes and degrade the reliability of predicted change masks. Robust change localization therefore requires that such spurious responses be suppressed while the structural integrity of change regions in complex, high-resolution scenes is maintained. We propose TriFusion-CD, a tri-branch framework that fuses complementary sources of information for reliable change localization. The first branch uses MobileSAM to provide global semantic guidance that promotes spatially coherent predictions. The second branch adopts the CLIP-ResNet50 image encoder with a change-aware enhancement module to extract detail-sensitive change features. The third branch performs frequency decomposition and interacts frequency features with CLIP text embeddings via cross-attention, producing a structural–semantic prior to suppress appearance-induced pseudo-changes. We further design a Semantic Attention Fusion Module (SAFM) to inject MobileSAM semantics into CLIP change features through cross-attention with learnable residual scaling. In addition, an Attention-Modulated Decoder (AMD) translates the fused guidance into multi-scale attention maps and performs progressive top-down refinement, extracting more spatially complete change regions. On the challenging SYSU-CD, JL1-CD, and CDD datasets, which exhibit diverse change patterns and frequent appearance-induced pseudo-changes, TriFusion-CD achieves 72.48% IoU/84.04% F1 on SYSU-CD, 66.04% IoU/79.54% F1 on JL1-CD, and 96.41% IoU/98.17% F1 on CDD, demonstrating strong performance. Full article

Home food safety (HFS) is a major contributor to foodborne illness, often originating in domestic settings. Although population-based studies using surveys, questionnaires, and interviews are commonly used to assess consumers’ HFS-related knowledge, attitudes, and practices (KAP), methodological heterogeneity limits comparability across studies. This scoping review aimed to map studies assessing consumers’ HFS-related KAP in high-income countries, describe recurrent methodological and reporting features, and identify areas of variability. Following the Arksey and O’Malley framework and JBI guidance, the literature published between 2000 and 2023 was systematically searched across five scientific databases, as well as governmental and institutional sources for the grey literature. Data extraction and synthesis were guided by an expanded 15-feature framework refined from a previous rapid review. A total of 274 documents were included (247 scientific articles and 27 governmental and institutional reports). Across the included studies, several methodological features showed high consistency, including primary data collection (93%), predominantly cross-sectional designs (91%), the use of closed-ended instruments (71%), quantitative analytical approaches (78%), and voluntary, non-incentivised participation (68%), suggesting the presence of a common descriptive methodological core. At the same time, substantial variability was observed in sample size (62%), study aims (52%), analytical strategies (52%), modes of administration (51%), geographic coverage (47%), thematic scope (44%), and study period (54%). The coexistence of methodological convergence and context-dependent variability poses challenges in terms of evidence synthesis and comparability in HFS-related KAP research. The 15-feature framework developed in this review provides a structured, non-prescriptive tool to support transparent description and comparison of methodological and reporting practices. By pinpointing common approaches and areas of divergence, this review offers a foundation for guiding future HFS-related KAP research and supporting the development of more comparable and policy-relevant evidence. Full article

Environmental, social, and governance (ESG) practices have become increasingly central to business sustainability strategies. Yet, the empirical literature remains fragmented regarding how ESG is translated into firm-level outcomes and sustainable value creation. This study conducts a scoping review to map the relationships among ESG practices, innovation, and organizational value creation, with particular attention to business sustainability. Reported in accordance with PRISMA 2020, with additional consideration of guidance specific to scoping reviews, searches in Scopus, Web of Science, and ScienceDirect identified 87 empirical studies. The review examines ESG conceptualization and measurement, the structural roles of innovation, and value-related outcomes. The findings reveal three dominant patterns: ESG is most often operationalized through rating-based indicators; innovation, especially green innovation and digital transformation, frequently acts as the mechanism through which ESG is translated into organizational change and performance outcomes; and value creation is increasingly assessed through both financial and sustainability-oriented indicators. Based on these findings, the study synthesizes recurring empirical patterns into an integrative sustainability framework in which ESG is interpreted as a strategic orientation, innovation as a capability conversion layer, and sustainable organizational value as the resulting outcome. Full article