Search Results (9,712)

Accurate and early identification of breast cancer from mammography is key to reducing breast cancer mortality, and automated analysis is challenging due to subtle lesion appearances, heterogeneous breast density, and the variance caused by modality. Standard Convolutional Neural Networks (CNNs) are excellent at capturing localized textures, whereas Vision Transformers (ViTs) capture long-range dependencies; however, both often struggle to produce a unified representation that consistently supports diagnostic decision-making. To address these limitations, this study presents a dual-stream framework integrating ConvNeXt for high-fidelity local feature extraction with Swin Transformer V2 for hierarchical global context modeling. A Bi-Directional Cross-Guidance (BDCG) mechanism is added to harmonize interactions between the two feature domains and ensure mutual information learning in the representations. Furthermore, a Prototype-Anchored Similarity Head (PASH) is used to stabilize classification using distance-based reasoning instead of using linear separation. Comprehensive experiments show the effectiveness of the proposed method using two benchmark datasets. On Dataset 1, the model achieves accuracy: 98.8%, precision: 98.7%, recall: 98.6%, and F1 score: 97.2%, outperforming existing models based on CNN, ViTs, and hybrid architectures, and provides a lower inference time (8.3 ms/image). On the more heterogeneous Dataset 2, the model maintains strong performance, with an accuracy of 97.0%, precision of 95.4%, recall of 94.8%, and F1-score of 95.1%, demonstrating its resilience to domain shift and imaging variability. These results underscore the value of structural multi-scale feature interaction and prototype-driven classification for robust mammographic analysis. The consistent performance across internal and external evaluations indicates the potential for the proposed framework to be reliably applied in computer-aided screening systems. Full article

Urban areas require context-specific bioenergy solutions to advance toward circular and sustainable energy systems. In Bogotá, urban pruning and grass-cutting residues constitute a relatively stable biomass stream; however, the absence of district-scale valorization infrastructure leads to their direct disposal in landfill. This study develops and applies a GIS-based planning methodology to support the territorial design of a small-scale anaerobic digestion plant using urban green waste. In this study, “small-scale” is understood as an early-stage urban facility concept compatible with the available pruning stream of approximately 1200–1300 t/month of valorizable biomass, corresponding only to an order-of-magnitude energy range of a few hundred kW${}_e$/kW${}_t$, rather than to a final engineering design. The approach integrates official geospatial data with logistical, environmental, and institutional criteria to characterize biomass availability and evaluate location alternatives under real urban constraints. A continuous location model based on the Weber problem is first applied to estimate a theoretical lower bound of spatial effort, using public schools weighted by enrollment as a proxy for sensitive urban demand. Subsequently, a GIS-assisted Analytic Hierarchy Process (AHP) is implemented to incorporate environmental exclusions, territorial compatibility, and the operational structure of exclusive waste service areas. Results show that the optimal geometric location diverges from the territorially feasible alternative once environmental restrictions and biomass supply coherence are explicitly considered. The findings highlight that urban bioenergy infrastructure planning is governed less by pure spatial efficiency than by the integration of supply, demand, and institutional constraints. The proposed methodology provides a reproducible decision-support tool for urban bioenergy planning and contributes to sustainable waste management, circular economy strategies, and local energy resilience in cities of the Global South. Full article

Objectives: To synthesize evidence on artificial intelligence (AI)-enabled medical history taking (anamnesis)—beyond large language models (LLMs) alone—and to translate findings into implications and research priorities for head and neck surgery. Methods: We performed a PRISMA-informed narrative review. Searches from database inception to 31 December 2025 (updated 3 January 2026) were conducted in MEDLINE (PubMed), Embase, Scopus, Web of Science Core Collection, IEEE Xplore, and ACM Digital Library, supplemented by medRxiv/arXiv screening and citation chasing. We included studies evaluating or describing AI-supported history capture/summarization, conversational interviewing, symptom checker/digital triage, EHR-integrated intake-to-decision support pipelines, voice interviewing, education/training systems, and governance/ethical considerations related to digital anamnesis. Findings were synthesized by system category and by cross-cutting outcome domains, with a head and neck surgery interpretive lens. Results: Fifty studies (2014–2025) were included. Evidence most consistently suggested feasibility and acceptability of pre-consultation computer-assisted history taking and the potential to reduce documentation burden and improve structured capture. In contrast, symptom checkers and digital triage tools showed highly variable diagnostic/triage performance and prominent safety concerns, highlighting the importance of conservative red-flag escalation strategies, continuous monitoring, and clear accountability. LLM-based diagnostic dialogue demonstrated strong performance in controlled evaluations, but prospective real-world validation, governance, and workflow integration remain limited. Conclusions: AI-enabled anamnesis comprises heterogeneous tools with uneven evidence. For head and neck surgery, potential near-term applications may include structured pre-visit intake, clinician-facing summarization, and training applications, whereas autonomous triage warrants harm-oriented, specialty-calibrated validation and robust governance prior to broader clinical reliance. Full article

This paper describes a new methodology implemented in the ESTE decision support system for evaluating the source term resulting from a nuclear weapon detonation. The methodology is based on a model of a stabilized radioactive mushroom cloud, parameterized as the source term for a Lagrangian particle dispersion model. It includes radionuclide composition, spatial distribution of aerosol and gaseous particles, and particle size distribution. This method is designed for rapid assessment of radiological impacts primarily at medium- and long-range distances, for example, in neighboring countries. The parametrization has been calibrated and adjusted using data from historical nuclear tests, and its performance is evaluated in terms of impacted area, range, and spatial overlap of fallout regions. A comparison is presented between ESTE calculations and field measurements obtained after the British nuclear tests conducted in the 1950s at the Maralinga Range (Australia), using historical ERA5 meteorological reanalyses from ECMWF. Full article

Accurate and early prediction of crop yield at the sub-field scale is essential for precision-agriculture and food-system planning. This study evaluates a phenology-based machine learning framework for winter wheat yield prediction using Sentinel-2 satellite imagery, climate reanalysis data, and field-level yield data. Phenological metrics derived from the normalised difference vegetation index (NDVI), the normalised difference water index (NDWI), and the normalised difference red-edge index (NDRE) were combined with accumulated seasonal rainfall and seasonal potential evapotranspiration, and multiple modelling strategies were assessed using a leave-one-field-out cross-validation (LOFO CV) scheme to ensure spatial generalisation. Among the evaluated models, the Random Forest (RF) algorithm achieved the highest overall performance, explaining up to 73% of the yield variability with a root mean square error (RMSE) of 0.88 t ha⁻¹ at optimal prediction timing (day of year 160–175). Integrating phenological and climatic covariates consistently improved prediction accuracy compared to models based only on phenological variables, while the inclusion of soil properties provided limited additional benefit at the examined spatial scale. Phenological metrics based on red-edge data, particularly the maximum NDRE, were the most influential predictors, highlighting the added value of red-edge spectral information beyond traditional red–near-infrared indices. Uncertainty analysis revealed spatially heterogeneous prediction uncertainty, particularly near field boundaries and in areas of complex spatial patterns. Overall, the proposed framework enables robust, early, and interpretable yield prediction at the sub-field scale, supporting uncertainty-aware decision-making in precision agriculture and offering a scalable foundation for regional crop monitoring. Full article

Artificial intelligence (AI) has shown substantial potential to improve patient outcomes in parenteral nutrition by enabling individualised nutritional strategies, early prediction of metabolic and infectious complications, and optimised real-time clinical decision-making. Evidence from global clinical practice demonstrates that AI integration can enhance patient safety, reduce complication rates, and improve resource utilisation. In Bulgaria, recent developments in parenteral nutrition reflect progress toward standardisation, wider availability of modern formulations, and alignment with international clinical guidelines. However, the adoption of AI-driven systems for personalised nutrition planning and continuous risk assessment remains limited. Key barriers include the availability and quality of clinical data, regulatory and ethical considerations, and the need for targeted training of healthcare professionals. This review highlights both the opportunities and challenges associated with implementing AI in parenteral nutrition in the Bulgarian context. Potential benefits include improved patient outcomes, shorter hospital stays, more efficient healthcare delivery, and alignment with international best practices. At the same time, overcoming infrastructural, regulatory, and educational barriers is essential for successful implementation. Conclusions: The integration of AI into parenteral nutrition requires a multidisciplinary approach that combines clinical expertise, technological innovation, and supportive health policy. Such an approach offers the potential to sustainably enhance patient care in Bulgaria and position national practice in line with leading global standards. Full article

Against the backdrop of population aging, community parks are important spaces for older adults’ daily activities, and perceived safety plays a key role in shaping their use and spatial satisfaction. This study selected six typical community parks in central Beijing, constructed an indicator system for safety perception needs, and applied an analytical KANO–IPA (Integrated Kano and Importance-Performance Analysis) approach to identify the demand attributes and optimization priorities of safety elements. The results reveal a clear hierarchy in older adults’ safety perception needs. Basic environmental and facility safety factors, such as pavement conditions and facility reliability, function as must-be needs. Elements related to spatial visibility, circulation, lighting, and wayfinding act as one-dimensional needs that steadily influence satisfaction, whereas features including natural surveillance, spatial enclosure, and activity atmosphere mainly enhance spatial experience as attractive needs. Priority analysis further indicates that circulation conditions and facility safety constitute the most critical aspects for improvement. Overall, older adults’ safety perception in community parks results from the combined effects of multiple spatial factors. Hierarchical spatial optimization can enhance user experience and improve resource allocation efficiency. The findings provide theoretical support and decision-making guidance for safety-oriented planning and age-friendly renewal of urban community parks in aging societies. Full article

Freshwater ecosystems play an important role in human survival, ecosystem functioning, and biodiversity conservation, yet industrialisation and urbanisation dump over 80% of untreated sewage into them. This inadequate wastewater management leads to enteric pathogens like Escherichia coli, Salmonella, Shigella, Campylobacter, Vibrio cholerae, Pseudomonas aeruginosa, and Legionella pneumophila that are responsible for a wide range of waterborne human diseases globally with extensive morbidity and mortality. The World Health Organization (WHO) estimates that at least 2 billion individuals drink water contaminated with pathogens, resulting in illnesses like cholera, dysentery, and diarrhoea, and approximately 50,000 diarrheal deaths annually. Classical epidemiology approaches are the basis for determining disease burden in public health, but they are limited in their capacity to predict future health risks. Quantitative microbial risk assessment (QMRA) addresses this by estimating the potential health risks of any exposure to microbial pathogens in any environment using four key elements, which include the identification of the microbial hazards, human exposure to the hazard through diverse activities, dose–response relationships, and the estimated risk of the infection. This review summarises information on freshwater pathogens, their occurrence, sources and health implications. The methodological approaches of QMRA in freshwater systems are reviewed with examples drawn from recreational activities, drinking water, and wastewater-impacted environments. Global QMRA studies indicate a wide range of infection risk estimates, reflecting differences in water sources, pathogens, and exposure conditions. Thus, QMRA is known to be a valuable public health tool for freshwater ecosystems, linking microbial contamination dynamics to health risk estimates that support proactive management and policy-relevant decision-making. Full article

Scarcity of healthcare resources requires prioritisation decisions that raise complex ethical, economic, and social challenges. While normative frameworks provide guidance on how such decisions ought to be made, growing evidence suggests that individuals differ substantially in how they approach morally charged allocation choices. This study investigates heterogeneity in decision-making styles and support for healthcare prioritisation criteria using an interdisciplinary approach that integrates health economics, social psychology, and computational methods to identify latent decision-making profiles among a sample of adults residing in Portugal. Data were collected from adults residing in Portugal using a structured online questionnaire comprising socio-demographic characteristics, decision-making styles, and preferences elicited through twenty hypothetical healthcare rationing scenarios. The results reveal three meaningful decision-making profiles characterised by different combinations of cognitive styles and ethical prioritisation patterns: analytically oriented decision-makers prioritising health gains; intuitive, context-sensitive decision-makers balancing clinical and social criteria; heuristic-driven decision-makers relying on simpler or less differentiated heuristics. These findings demonstrate that, within this sample, healthcare prioritisation preferences are shaped by systematic variations in decision style rather than a single moral or rational framework. By linking behavioural heterogeneity with ethical decision-making, this study contributes to theoretical debates on healthcare rationing and demonstrates the value of clustering techniques for uncovering latent structures in complex decision data. The results provide insights relevant for the design of decision-support systems and rationing policies, which may be adapted to accommodate heterogeneous decision styles in comparable settings. Full article

Precision Agriculture (PA) is increasingly applied to nutraceutical cropping systems, where agronomic productivity must be integrated with the stabilization of phytochemical quality and environmental sustainability. This structured narrative review synthesizes scientific evidence (primarily 2010–2025) on the use of Unmanned Aerial Vehicle (UAV)-based multispectral and thermal sensing, LiDAR-derived canopy characterization, Internet of Things (IoT) monitoring, and artificial intelligence (AI)-driven analytics in medicinal, aromatic, and functional crops. The literature indicates that PA enhances high-resolution monitoring of crop–environment interactions, supporting site-specific irrigation, nutrient management, and stress detection. Under validated conditions, these interventions are associated with improved yield stability, resource-use efficiency, and modulation of secondary metabolite accumulation. However, reported outcomes vary substantially across species, agroecological contexts, and experimental scales, and most studies remain plot-scale or pilot-scale, limiting large-scale generalization. Moringa oleifera Lam. is examined as a model species for Mediterranean and semi-arid systems. Evidence suggests that integrated spectral, structural, and environmental monitoring can support optimized irrigation scheduling, canopy uniformity, and phytochemical consistency. Nonetheless, genotype-specific calibration, multi-season validation, standardized metabolomic benchmarking, and cross-regional transferability remain significant research gaps. Overall, PA represents a scientifically promising but still maturing framework for nutraceutical agriculture. Future progress will require rigorous multi-site validation, improved model robustness, standardized sustainability metrics, and comprehensive economic assessments to ensure scalability and long-term impact. Full article

Background: Left ventriculo-arterial coupling (VAC) integrates the interaction between left ventricular contractility and the arterial system, representing a key determinant of cardiovascular efficiency. In rapidly changing hemodynamic states such as septic or cardiogenic shock, conventional indices of pressure or flow alone may be misleading. VAC provides a unified physiological framework to assess global cardiovascular performance and guide therapy. Objective: To review the physiological foundations, bedside assessment, and therapeutic applications of VAC in critically ill patients with rapidly fluctuating circulatory conditions. Methods and Content: The article revisits the underlying principles of VAC, expressed as the ratio between arterial elastance (Ea) and end-systolic elastance (Ees), and discusses their derivation from the pressure–volume relationship. Practical echocardiographic methods for bedside estimation, including the non-invasive single-beat approach, are outlined with illustrative figures. The review further examines how VAC patterns evolve in sepsis, cardiogenic shock, and heart failure and how this integrative index clarifies paradoxical responses to vasoactive and inotropic therapies. Specific therapeutic phenotypes are proposed according to Ea/Ees profiles, providing a structured approach to optimise coupling and restore circulatory efficiency. Summary: VAC offers a physiology-based perspective on cardiovascular performance, enabling clinicians to interpret complex hemodynamic changes beyond traditional measures of ejection fraction or mean arterial pressure. Its dynamic tracking may refine the assessment of therapeutic trajectories and improve bedside decision-making. Conclusions: By integrating ventricular and arterial function into a single measure, VAC bridges cardiovascular physiology and clinical practice. Its incorporation into routine critical care monitoring could enhance individualised hemodynamic management and serve as a foundation for future outcome-driven studies. Methodology: This narrative review was conducted using a structured literature search to ensure comprehensive coverage of contemporary evidence regarding ventriculo-arterial coupling (VAC) in critical care and shock states. A systematic search of PubMed/MEDLINE, Embase, and Scopus databases was performed from database inception through October 2025. The following key search terms were used: “ventriculo-arterial coupling”; “arterial elastance”; “end-systolic elastance”; “Ea/Ees”; “pressure–volume loops”; “septic shock”; “cardiogenic shock”; “critical care echocardiography”; “point-of-care ultrasound”; “mechanical circulatory support”. Reference lists of relevant articles, review papers, and consensus documents were also manually screened to identify additional pertinent studies. Only English-language publications were included. Both seminal foundational studies and recent contemporary investigations were reviewed to provide historical context and up-to-date clinical applicability. Full article

Quantum computing offers potential computational advantages, yet its integration into autonomous decision-making systems remains largely unexplored. This paper addresses the need for a unified framework that systematically combines quantum computation with agent-based artificial intelligence. We examine how quantum technologies can enhance the capabilities of autonomous agents and, conversely, how agentic AI can support the advancement of quantum systems. We analyze both directions of this synergy and present conceptual and technical foundations for future quantum–agentic platforms. Our work introduces a formal definition of quantum agents and outlines architectures that integrate quantum computing with agent-based systems. As concrete proof-of-concept implementations, we develop and evaluate three quantum agent prototypes: (i) a Grover-based decision agent for quantum search-driven action selection, (ii) a variational quantum reinforcement learning agent for adaptive policy learning in a multi-armed bandit setting, and (iii) an adaptive quantum image encryption agent that autonomously selects encryption strategies based on entropy-driven feedback. These prototypes demonstrate practical realizations of quantum agency in decision-making, learning, and security contexts under NISQ-era constraints. Furthermore, we discuss application domains including quantum-enhanced optimization, hybrid quantum–classical orchestration, autonomous quantum workflow management, and secure quantum information processing. By bridging these fields, we introduce a structured theoretical and architectural framework for quantum–agentic systems, providing formal definitions, system models, and early operational prototypes that illustrate the feasibility of quantum-enhanced agency under NISQ constraints. Full article

This study assesses the extent to which an AI-driven circular waste management tool, previously developed by the same authors as a decision-support system for the circular management of healthcare waste in compliance with international guidelines, reflects the operational needs and perceived priorities of healthcare professionals and environmental managers. Within a context characterised by high regulatory complexity and increasing pressure toward more sustainable management models, the research adopts a qualitative approach based on the thematic analysis of 11 semi-structured interviews, followed by a systematic mapping of the emergent themes onto the tool’s thematic areas, indicators, and operational actions. The results demonstrate a high degree of alignment between the tool and operational practice, with 93% of the tool’s actions supported by empirical evidence and the emergence of a shared core cluster focused on hard-to-manage waste streams, mandatory training, and day-to-day operational challenges. The alignment between the priorities expressed by interviewees and the importance scores generated by the computational model is high for actions of greater relevance, while it decreases for less frequent actions that are more context-dependent. Circular economy practices are recognised as relevant but remain predominantly positioned at a strategic rather than an operational level. Overall, the study confirms the conceptual robustness of the tool and identifies its main limitations and the conditions required for its integration into hospital workflows. Full article

The purpose of this study is to examine whether analyst-level credit can be assigned quantitatively in a lightweight human-feedback decision-support pipeline. In intelligence and national security workflows, analysts often provide edits, comments, and evaluative feedback during the production of analytic products, yet these intermediate contributions are usually discarded, leaving no auditable record of how individual feedback shaped the final output. To address this problem, this study proposes a proof-of-concept Analyst-of-Record framework that combines synthetic analyst feedback, a linear ridge reward model, first-order influence functions, and additive Shapley aggregation to estimate both feedback-item and analyst-level contribution scores. The research design uses the Fact Extraction and VERification (FEVER) fact-verification dataset under controlled experimental settings. The pipeline retrieves evidence with Best Matching 25 (BM25), generates a grounded template-based response, derives three synthetic analyst feedback channels from FEVER annotations, trains a reward model on simple claim–answer and analyst-identity features, and aggregates per-feedback influence scores into an Analyst Contribution Index (ACI). The main experiments are conducted on a 500-claim subset across five random seeds, with additional ablation and bootstrap analyses used to assess sensitivity and stability. The findings show that the reward model achieves a mean validation $R^2$ of $0.801\pm 0.037$, indicating that the synthetic feedback signals are learnable under the selected featureization. The analyst-level contribution scores remain stable across random seeds, with approximately half of the total influence magnitude attributed to the explanation-quality channel and the remainder split across the other two channels. Ablation results further show that removing the explanation-quality channel collapses validation fit, while bootstrap resampling demonstrates tight concentration of absolute ACI magnitudes. Theoretically, this study extends attribution research beyond document-only grounding by showing how analyst feedback itself can be modeled as an object of contribution analysis. It also demonstrates that influence functions and Shapley-style aggregation can be adapted into a tractable framework for estimating interpretable analyst-level credit in a reproducible experimental setting. Practically, the proposed framework offers an initial foundation for more traceable and accountable decision-support workflows in which intermediate analyst contributions can be preserved rather than lost. The results also provide a feasible implementation path for future systems that incorporate stronger generators, richer evidence representations, and real analyst annotations. Full article

Background: Congenital scoliosis (CS) associated with multiple vertebral anomalies (MVAs) represents a biologically dynamic deformity in which cumulative segmental asymmetry, residual growth potential, and mechanobiological modulation interact to drive progression. Unlike isolated congenital lesions, MVAs exhibit growth-dependent and configuration-specific behavior, complicating risk stratification and timing of intervention. Despite extensive literature on congenital deformities, an integrated growth-oriented decision framework for this subgroup remains lacking. Methods: This narrative review synthesizes embryological, biomechanical, and clinical evidence related to vertebral growth potential, anomaly configuration, progression patterns, and age-dependent treatment strategies in CS with MVAs. A structured literature search of major databases was performed, and findings were analyzed thematically to propose a biologically grounded growth-based decision framework. Results: Across the literature, three interdependent determinants of progression consistently emerge: anomaly configuration, residual segmental growth capacity, and mechanobiological amplification during growth. High-risk configurations—particularly mixed formation–segmentation defects and fully segmented hemivertebrae with contralateral growth arrest—demonstrate rapid and often non-linear progression. Thoracic involvement further modifies clinical urgency due to its impact on pulmonary development. Integration of developmental biology and mechanobiological principles supports a structured, growth-informed approach to surveillance and intervention timing. Conclusions: MVAs should be conceptualized as dynamic growth systems rather than static structural defects. A shift from angle-driven to growth-informed decision-making may enhance early identification of high-risk patterns while minimizing unnecessary premature fusion in lower-risk cases. Adoption of a structured growth-based framework provides a biologically coherent foundation for individualized management and long-term optimization of spinal and thoracic development. Full article