Search Results (823)

Elderly patients often experience multimorbidity and long-term polypharmacy, making medication safety a critical challenge in disease management. In China, the concurrent use of Western medicines and proprietary Chinese medicines (PCMs) further complicates this issue, as potential drug interactions are often implicit, increasing risks for physiologically vulnerable older adults. Although large language model-based medical question-answering systems have been widely adopted, they remain prone to unsafe outputs in medication-related contexts. Existing retrieval-augmented generation (RAG) frameworks typically rely on static retrieval strategies, limiting their ability to appropriately allocate retrieval and verification efforts across different question types. This paper proposes FusionGraphRAG, an adaptive RAG framework for geriatric disease management. The framework employs query classification-based routing to distinguish questions by complexity and medication relevance; integrates dual-granularity knowledge alignment to connect fine-grained medical entities with higher-level contextual knowledge across diseases, medications, and lifestyle guidance; and incorporates explicit contradiction detection for high-risk medication scenarios. Experiments on the GeriatricHealthQA dataset (derived from the Huatuo corpus) indicate that FusionGraphRAG achieves a Safety Recall of 71.7%. Comparative analysis demonstrates that the framework improves retrieval accuracy and risk interception capabilities compared to existing graph-enhanced baselines, particularly in identifying implicit pharmacological conflicts. The results indicate that the framework supports more reliable geriatric medical question answering while providing enhanced safety verification for medication-related reasoning. Full article

The research explores the educational potential of artistic mediation with children at risk of social exclusion, drawing on the narratives of twenty early years and primary school teachers. Using a qualitative, phenomenological approach, it examines perceptions of openness to the creative process, the use of art in teaching practice and its value as a socio-educational tool. The findings show that experiences of artistic mediation generate feelings of harmony, concentration and achievement, thus fostering a more collaborative and emotionally balanced classroom climate. The study also observes that art serves as a means for teachers’ reflective practice, encouraging critical thinking, the formulation of questions and an approach to assessment that focuses more on processes than products. In vulnerable contexts, artistic mediation proves particularly effective for expressing emotions, strengthening self-esteem and reinforcing group cohesion. Taken together, the findings suggest that artistic mediation should be understood beyond its instrumental value, recognising it as a transformative practice that promotes both educational inclusion and professional reflection on teaching, thereby helping to enhance the quality and humanistic purpose of pedagogical interventions. Full article

Background/Objective: Systematic reviews (SRs) with network meta-analysis (NMA) support evidence-based decision-making by enabling both direct and indirect comparisons across multiple interventions. Given the expanding use of Janus kinase (JAK) inhibitors in rheumatoid arthritis (RA), the methodological rigor of SRs with NMA is essential for trustworthy conclusions. This study is aimed at evaluating the methodological quality of SRs with NMA assessing the efficacy and/or safety of JAK inhibitors in RA. Methods: PubMed and Embase were searched for full-text SRs with NMAs evaluating JAK inhibitors as a therapeutic class in RA. Eligible publications were English-language articles reporting efficacy and/or safety outcomes. Narrative reviews, letters, duplicates, reviews focused on a single JAK inhibitor, and reviews without quantitative synthesis were excluded. Three independent reviewers assessed methodological quality using AMSTAR 2. Descriptive statistics were used to summarize findings. Results: Of the 222 records identified, 18 SRs with NMA met the inclusion criteria: 5 focused on efficacy, 5 on safety, and 8 assessed both. The most consistently fulfilled AMSTAR 2 items were a clearly defined PICO question (100%), duplicate study selection (100%), and reporting of conflicts of interest (100%). Common shortcomings included lack of protocol registration (44%), incomplete reporting of the search strategy (39%), and absence of publication bias assessment (50%). Risk-of-bias assessment varied by review focus: all safety reviews complied (100%), compared with 20% of efficacy reviews and 37% of mixed reviews. Conclusions: Most SRs with NMA of JAK inhibitors in RA present relevant methodological limitations, particularly in protocol registration, search reporting, and risk-of-bias assessment. Methodological standards were generally higher in safety-focused reviews, underscoring the need for more consistent and rigorous conduct and reporting, especially in efficacy and mixed reviews, to strengthen confidence in NMA-derived conclusions. Full article

Vision-language models (VLMs) integrate visual and textual information and are increasingly being used as innovative tools in educational applications. However, there is a lack of evidence regarding current practices for integrating VLMs into teaching and learning. To address this research gap and identify the opportunities and challenges associated with the integration of VLMs in education, this paper presents a systematic review of VLM use in formal educational contexts. Peer-reviewed articles published between 2020 and 2025 were retrieved from five major databases: ACM Digital Library, Scopus, Web of Science, Engineering Village, and IEEE Xplore. Following the PRISMA-guided framework, 42 articles were selected for inclusion. Data were extracted and analyzed against six research questions: (1) where VLMs are applied across academic disciplines and educational levels; (2) what types of VLM solutions are deployed and which image–text modalities they infer and generate; (3) the pedagogical roles of VLMs within teaching workflows; (4) reported outcomes and benefits for learners and instructors; (5) challenges and risks identified in practice, together with corresponding mitigation strategies; and (6) reported evaluation methods. The included studies span K-12 through higher education and cover diverse disciplines, with deployments dominated by pre-trained models and a smaller number of domain-adapted approaches. VLM-supported pedagogical functions cluster into five roles: analyst, assessor, content curator, simulator, and tutor. This review concludes by discussing implications for VLM adoption in educational settings and offering recommendations for future research. Full article

Background: Low peak bone mass (PBM) is a major contributor to later osteoporosis risk. This study sought to understand young people’s knowledge of factors associated with bone health. Methods: Young people in Aotearoa New Zealand were approached. Eight focus groups (26 participants in total, aged 11 to 17 years) were conducted using a semi-structured approach with open-ended questions and prompts. Transcripts were thematically coded using an inductive content analysis approach. Results: Knowledge of factors associated with good bone health was limited. There was a general awareness of the positive and negative impacts of many lifestyle behaviours on health generally, but not specifically PBM. Dairy intake was commonly mentioned as being beneficial for bone health. Some participants reported potential benefits of sport, but most did not know that weight bearing activity specifically was beneficial. Conclusions: Knowledge of osteoporosis and lifestyle factors that impact PBM was limited. Educational interventions involving promotion of bone health knowledge and supporting weight bearing physical activity in adolescents may be an important contributor to public health strategies. Full article

This study addresses a pivotal strategic issue in hospitality e-commerce: how hotels can optimize cooperation with heterogeneous online travel agencies (OTAs). Moving beyond the conventional question of whether to cooperate, we investigate the interrelated decisions of which OTA type to partner with (quality-focused vs. price-focused) and which business model to adopt (merchant vs. agency). We develop a game-theoretic model that incorporates key e-commerce factors, including hotel capacity constraints, cross-channel spillover effects, and differential consumer acceptance of OTA types. Our analysis yields a contingent decision framework. We demonstrate that OTA cooperation becomes beneficial only when a hotel’s room capacity exceeds its direct-channel demand. The optimal strategy evolves with capacity: hotels with moderate capacity should partner with a single OTA type—predominantly the quality-focused one—while larger hotels should engage both types to maximize market coverage. In terms of business models, smaller hotels benefit from the risk-shifting merchant model, whereas larger hotels capture higher margins through the agency model. A key finding is the general superiority of a differentiated approach: applying the agency model to quality-focused OTAs and the merchant model to price-focused OTAs. This research provides a structured analytical framework to guide hotel managers in crafting e-commerce platform strategies and offers scholars a foundation for further inquiry into platform competition and contract design in digital marketplaces. Full article

Explainability in deep learning (XDL) for healthcare is increasingly portrayed as essential for addressing the “black box” problem in clinical artificial intelligence. However, this universal transparency mandate may create unintended consequences, including cognitive overload, spurious confidence, and workflow disruption. This paper examines a fundamental question: Is explainability a panacea that resolves AI’s trust deficit, or a Pandora’s box that introduces new risks? Drawing on general systems theory we demonstrate that the answer is profoundly context dependent. Through systemic analysis of current XDL methods, Saliency Maps, LIME, SHAP, and attention mechanisms, we reveal systematic disconnects between technical transparency and clinical utility. This paper argues that XDL is a context-dependent systemic property rather than a universal requirement. It functions as a panacea when proportionately applied to high-stakes reasoning tasks (cancer treatment planning, complex diagnosis) within integrated socio-technical architectures. Conversely, it becomes a Pandora’s box when superficially imposed on routine operational functions (scheduling, preprocessing) or time-critical emergencies (e.g., cardiac arrest) where comprehensive explanation delays lifesaving intervention. The paper proposes a risk-stratified framework recognizing that a specific subset of healthcare AI applications—those involving high-stakes clinical reasoning—require comprehensive explainability, while other applications benefit from calibrated transparency appropriate to their clinical context. We conclude that explainability is neither a cure-all nor an inevitable harm, but rather a dynamic equilibrium requiring continuous rebalancing across technical, cognitive, and organizational dimensions. Full article

This article offers a comprehensive technical and mechanical review of the Tesla turbine, an innovative device conceived by Nikola Tesla. The core research question guiding this review is: How can the design and application of the Tesla turbine be optimized to overcome its current efficiency limitations and unlock its full potential across various energy recovery technologies? The analysis focuses on the mechanical design of the turbine, illustrating the configuration of co-axial discs without blades mounted on a central shaft, and on the fluid dynamic phenomena that generate torque through the viscous boundary layer between the discs. Mathematical models based on the equations of viscous motion and CFD simulations are used to evaluate mechanical and fluid-dynamic losses, such as viscous friction, edge losses, and inlet duct losses. The work describes mechanical engineering challenges related to efficiency and performance, highlighting optimization techniques for the number and spacing of the discs, nozzle geometry, and thermal management to mitigate the risk of overheating. Finally, potential application areas in microturbine technology for low-enthalpy thermal cycles and energy recovery are examined. The article makes a significant contribution to applied mechanical engineering, offering design guidelines and an updated overview of the challenges and opportunities of Tesla turbine technology. Full article

Although Large Language Models (LLMs) excel in language tasks, producing fluent and seemingly high-quality text, their outputs are essentially probabilistic predictions rather than verified facts, rendering reliability unguaranteed. This issue is particularly pronounced when models lack the required knowledge, which significantly increases the risk of fabrications and misleading content. Therefore, understanding whether LLMs know when they lack knowledge is of critical importance. This work systematically evaluates leading LLMs on their ability to recognize knowledge insufficiency and examines several training-free techniques to foster this metacognitive capability, referred to as “integrity” throughout this research. For rigorous evaluation, this study firstly develops a new Question-Answering (Q&A) dataset called Honesty. Specifically, events emerging after the model’s deployment are utilized to generate “unknown questions,” ensuring they fall outside LLMs’ knowledge boundaries, while “known questions” are drawn from existing Q&A datasets, together constituting the Honesty dataset. Subsequently, based on this dataset, systematic experiments are conducted using multiple representative LLMs (e.g., GPT-4o and DeepSeek-V3). The results reveal that semantic understanding and reasoning capabilities are the core factors influencing “integrity.” Furthermore, we find that well-crafted prompts markedly improve models’ integrity, and integrating them with probability- or consistency-based uncertainty evaluation methods yields even stronger performance. These findings highlight the considerable potential of LLMs to express uncertainty when they lack knowledge, and we hope these observations can lay the groundwork for developing more reliable models. Full article

Background: Chronic diseases such as type 2 diabetes mellitus and cardiovascular disease and their cardiometabolic risk factors require management, which includes lifestyle interventions. Rural and remote residents are disproportionately affected by these conditions compared to their urban counterparts. Studies have examined barriers to chronic disease and cardiometabolic risk factor management in urban environments, but rural perspectives remain underexplored, especially in Western Australia (WA) with its vast geography. This study examined rural general practitioners’ (GPs) views on barriers and facilitators to chronic disease and cardiometabolic care in rural WA through lifestyle management. Methods: This qualitative study used semi-structured interviews with 15 rural WA GPs recruited via rural networks using convenience and snowball sampling. Braun and Clarke’s reflexive thematic analysis was used to identify patterns and themes within the qualitative data that addressed the study questions. Results: According to rural general practitioners, major barriers to chronic disease and cardiometabolic risk care included geographic isolation, socioeconomic disadvantage and an obesogenic food environment in rural areas, as well as severe time and financial constraints for GPs and workforce shortages with a high turnover and lack of accessible allied health professionals. Facilitators included co-located multidisciplinary teams, case management/health coaching, better remuneration for complex consultations involving preventive care and upstream policy measures, such as improving healthy food affordability and availability. Conclusion: Rural patients face systemic, geographic and socioeconomic barriers that are substantially greater than those in urban settings; these barriers impact GPs caring for their patients with chronic disease and cardiometabolic risk factors. Targeted solutions to these barriers such as attention to workforce issues, investment in lifestyle coaching approaches and having dedicated case managers, could reduce rural–urban inequities in chronic disease outcomes. Full article

Seconds count differently for people in danger. We present a real-time streaming pipeline for audio-based detection of hazardous life events affecting life and property. The system operates online rather than as a retrospective analysis tool. Its objective is to reduce the latency between the occurrence of a crime, conflict, or accident and the corresponding response by authorities. The key idea is to map reality as perceived by audio into a written story and question the text via a large language model. The method integrates streaming, zero-shot algorithms in an online decoding mode that convert sound into short, interpretable tokens, which are processed by a lightweight language model. CLAP text–audio prompting identifies agitation, panic, and distress cues, combined with conversational dynamics derived from speaker diarization. Lexical information is obtained through streaming automatic speech recognition, while general audio events are detected by a streaming version of Audio Spectrogram Transformer tagger. Prosodic features are incorporated using pitch- and energy-based rules derived from robust F0 tracking and periodicity measures. The system uses a large language model configured for online decoding and outputs binary (YES/NO) life-threatening risk decisions every two seconds, along with a brief justification and a final session-level verdict. The system emphasizes interpretability and accountability. We evaluate it on a subset of the X-Violence dataset, comprising only real-world videos. We release code, prompts, decision policies, evaluation splits, and example logs to enable the community to replicate, critique, and extend our blueprint. Full article

This paper reports a design-rationale account of building and deploying a small ecosystem of AI-driven educational conversational agents with distinct pedagogical personas. Two strands target school contexts: (i) Talk to Bill, a historically grounded Shakespeare interlocutor intended to support close reading, contextual understanding, and interpretive dialogue; and (ii) Here to Help, a set of UK GCSE subject- and exam-board-specific tutors designed for formative practice in recognised question formats with feedback and iterative improvement. The third strand comprises six complementary assistants for an undergraduate Human–Computer Interaction (HCI) module, each bounded to a workflow-aligned role (e.g., empathise-stage coaching, study planning, course operations), with guardrails to privilege process quality over answer generation. We describe how persona differentiation is mapped to established learning, engagement, and motivation theories; how retrieval-augmented generation and provenance cues are used to reduce hallucination risk; and what early deployment observations suggest about orchestration, integration, and incentives. The contribution is a transferable, auditable rationale linking theory to concrete dialogue and UI moves for multi-persona tutoring ecosystems, rather than a claim of causal learning gains. Full article

The oral microbiome has become an emerging focus of oral cancer research, with growing evidence linking microbial communities to disease development, progression, and prognosis. However, there is limited consensus on optimal sampling strategies, storage methods, and analytical approaches. This narrative review critically evaluates current strategies for sampling, preservation, DNA extraction, sequencing, and data analysis in oral microbiome research related to oral cancer. We compared commonly used sampling methods, including saliva, oral rinse, swab, brush, and tissue biopsy, and reviewed preservation conditions, extraction kits, sequencing platforms, and analytical pipelines reported in recent oral microbiome studies. Sampling approaches affect microbial yield and site specificity. Saliva and oral rinse samples are convenient and noninvasive but may dilute lesion-specific microbial signals, whereas lesion-directed swabbing or brushing yields greater microbial biomass and biological relevance. Preservation media and storage temperature significantly influence microbial stability, and DNA extraction methods vary in their ability to remove host DNA. Although 16S rRNA gene sequencing remains the most common approach, shotgun metagenomics offers higher resolution and function insights but is still limited by clinical applicability. Differences in data pre- and post-processing models and normalization strategies further contribute to inconsistent microbial profiles. Given that oral mucosal sites differ markedly in structure and microenvironment, careful consideration is required to ensure that collected samples accurately represent the biological question being addressed. Methodological consistency across all workflow stages—from collection to analysis—is essential to generate reproducible, high-quality data and to enable reliable translation of oral microbiome research into clinical applications for cancer detection and risk assessment. Together, these insights provide a framework to guide future study design and support the development of clinically applicable microbiome-based biomarkers. Full article

Interest in general AI is widespread, and much is expected from its large-scale adoption in the healthcare sector. However, the success of uHealth implementations relies on genuine trust, beyond technical performance. Neuroception of psychological safety (NPS), grounded in polyvagal theory, encompasses the human subconscious and automatic processes of safety and risk detection. We conducted a cross-sectional survey to explore a hypothetical connection between NPS and the perception of general AI in the uHealth context, by an anonymous online questionnaire comprising the following: Neuroception of Psychological Safety Scale (NPSS), four-item AI Attitude Scale (AIAS-4), and questions on AI threat, age, gender, and level of education. Multivariate analysis was performed using covariance-based structural equation modeling. We received 201 responses: 73 (36.3%) males vs. 128 (63.7%) females, all adults with varying levels of education (from 0 = basic formal education to 4 = master’s degree). Respondents belonged to four demographic cohorts: from Baby boomers to Generation Z. SEM results indicated that attitudes towards AI-driven health interventions are significantly impacted by social engagement and compassion (NPSS factors). Gender, education, and demographic cohort were confirmed as significant covariates. NPS-related attitudes towards AI should be considered and analyzed by healthcare providers, application developers, and policy or regulatory authorities. Full article

The pervasive deployment of large language models (LLMs) has given rise to mounting concerns regarding the safety and security of the content generated by these models. Nevertheless, the absence of comprehensive evaluation methods constitutes a substantial obstacle to the effective assessment and enhancement of the safety and security of LLMs. In this paper, we develop the Safety and Security (S&S) Benchmark, integrating multi-source data to ensure comprehensive evaluation. The benchmark comprises 44,872 questions covering ten major risk categories and 76 fine-grained risk points, including high-risk dimensions such as malicious content generation and jailbreak attacks. In addition, this paper introduces an automated evaluation framework based on multi-model judgment. Experimental results demonstrate that this mechanism significantly improves both accuracy and reliability: compared with single-model judgment (GPT-4o, 0.973 accuracy), the proposed multi-model framework achieves 0.986 accuracy while maintaining a similar evaluation time (~1 h) and exhibits strong consistency with expert annotations. Furthermore, adversarial robustness experiments show that our synthesized attack data effectively increases the attack success rate across multiple LLMs, such as from 14.76% to 27.60% on GPT-4o and from 18.24% to 30.35% on Qwen-2.5-7B-Instruct, indicating improved sensitivity to security risks. The proposed unified scoring metric system enables comprehensive model comparison; summarized ranking results show that GPT-4o achieves consistently high scores across ten safety and security dimensions (e.g., 96.26 in ELR, 97.63 in PSI), while competitive open-source models such as Qwen2.5-72B-Instruct and DeepSeek-V3 also achieve strong performance (e.g., 96.70 and 97.63 in PSI, respectively). Although all models demonstrate strong alignment in the safety dimension, they exhibit pronounced weaknesses in security—particularly against jailbreak and adversarial attacks—highlighting critical vulnerabilities and providing actionable direction for future model hardening. This work provides a comprehensive, scalable solution and high-quality data support for automated evaluation of LLMs. Full article