Search Results (1,084)

Accurate flood detection is critical for disaster response, yet the scarcity of diverse annotated datasets hinders robust model development. Existing resources typically suffer from limited geographic scope and insufficient annotation granularity, restricting the generalization capabilities of computer vision methods. To bridge this gap, we introduce AIFloodSense, a comprehensive evaluation benchmark designed to advance domain-generalized Artificial Intelligence for climate resilience. The dataset comprises 470 high-resolution aerial images capturing 230 distinct flood events across 64 countries and six continents. Unlike prior benchmarks, AIFloodSense ensures exceptional global diversity and temporal relevance (2022–2024), supporting three complementary tasks: (i) Image Classification, featuring novel sub-tasks for environment type, camera angle, and continent recognition; (ii) Semantic Segmentation, providing precise pixel-level masks for flood, sky, buildings, and background; and (iii) Visual Question Answering (VQA), enabling natural language reasoning for disaster assessment. We provide baseline benchmarks for all tasks using state-of-the-art architectures, demonstrating the dataset’s complexity and its utility in fostering robust AI tools for environmental monitoring. Crucially, we show that despite its compact size, AIFloodSense enables better generalization on external test sets than much larger alternatives, validating the premise that rigorous diversity is more effective than scale for training robust flood detection models, and is made publicly available to accelerate further research in the field. Full article

Background: Organisational health literacy (OHL) is increasingly recognised as a system-level strategy to address health literacy-related inequities in healthcare, yet evaluation of practical OHL tools and frameworks remain limited. This study aimed to examine the implementation experiences of the Danish OS! to inform refinements, and to examine the feasibility of the refined version, renamed OHL-Act, in practice. Methods: A two-phase study guided by the RE-AIM framework was conducted. Phase 1 comprised a formative evaluation of OS! based on interviews from previous applications, informing refinement. Phase 2 involved feasibility testing of OHL-Act in a specialised diabetes centre. Results: Across implementing organisations, OS! was experienced as a practical approach supporting reflection and the generation of OHL improvement ideas, while also revealing barriers. These insights informed refinements, including clearer language, more structured facilitation guidance, and explicit prompts addressing health literacy challenges and high-risk situations. Feasibility findings indicated that OHL-Act could be delivered as intended and was perceived as acceptable, relevant, and useful in supporting reflection and the generation of OHL improvement ideas. Conclusions: OHL-Act represents a structured, co-creational approach to support OHL work. Further research is needed to examine how generated improvement ideas translate into sustained action and their potential implications for equity. Full article

Accurate prediction of thermo-mechanical behavior in Fused Deposition Modeling (FDM) is often limited by mismatches between idealized Computer-Aided Design (CAD) geometry and path-dependent material deposition. This paper presents a G-code-driven, filament-level modeling and process-simulation workflow for complex geometries and infill strategies, especially toolpaths with in-plane inclinations. Extrusion segments are parsed from slicing G-code to obtain endpoints and process parameters, and each filament is reconstructed as a path-aligned rectangular bead using a dedicated local coordinate system. Progressive deposition is simulated in ANSYS Parametric Design Language (APDL) via an element birth–death method, enhanced by a centroid-based element selection strategy that reduces dependence on strictly aligned hexahedral partitions and improves robustness for complex meshes. A nonlinear transient thermal analysis is performed, and temperatures are mapped to the structural model through an indirect thermo-mechanical coupling scheme to predict warpage and residual stresses. Case studies on square plates with triangular and hexagonal infills (with/without sidewalls and a bottom base) show that the high-temperature zone follows newly deposited paths with peak temperatures near 220 °C, while displacement and von Mises stress accumulate and are strongly affected by infill topology and boundary conditions. Full article

Traditional approaches to university-level history education often fail to provide immersive and interactive environments that foster deep cognitive engagement. To address these limitations, we developed an AI-enabled virtual museum system that integrates AI-generated content with knowledge graphs through a multi-layered architecture. The system architecture follows a three-tier framework: a front-end interaction layer (Unity/Unreal Engine) for real-time user engagement, a core service layer for intelligent event scheduling and response control (Chat General Language Model/Stable Diffusion), and a data and model layer (My Structured Query Language/MongoDB) to provide structured knowledge. To evaluate the system’s effectiveness, a four-week controlled experiment was conducted with 83 university students. The experimental group using the AI virtual museum showed a significantly higher mean post-test score (84.5 ± 6.8) than that of the control group (71.6 ± 7.9), with statistical significance at p < 0.001, starting from nearly identical baseline scores (61.2 and 60.4 for the experimental and control groups). Correlation analysis was conducted to identify scenario simulations (r = 0.59) and deep inquiry tasks (r = 0.54) as key drivers of learning mastery. By aligning advanced system engineering with educational theory, the results of this study offer a solution for high-fidelity, intelligent digital educational platforms, proposing a validated model for integrated system innovation in education. Full article

Global forced displacement has reached unprecedented levels, with more than 123 million people uprooted by the end of 2024. Although older adults represent a growing proportion of refugee populations, their mental health needs remain overlooked. This scoping review synthesized current evidence on depression among older adult refugees aged 50 years and older. Guided by the Joanna Briggs Institute methodology and reported using PRISMA-ScR standards, searches were conducted in CINAHL, PsycINFO, AgeLine, and Medline for English-language publications from 2015 to 2025. A total of 1971 records were identified, with nine studies (N = 1370 participants) meeting eligibility criteria. Most studies employed cross-sectional designs and were conducted in high-income countries. Depression prevalence was consistently elevated, with rates ranging from 22% to over 70%, depending on population and measurement tools. Risk factors included female sex, widowhood, low socioeconomic status, chronic illness, functional impairment, trauma exposure, language barriers, social isolation, and limited access to care. Protective influences such as family support, higher socioeconomic status, and improved living conditions were identified but inconsistently reported. Findings indicate that older refugees are at high risk of depression, often shaped by intersecting aging- and displacement-related vulnerabilities. Findings highlight the need for culturally specific tools and longitudinal research to inform culturally safe care for older refugees. Full article

Background/Objectives: Age-related hearing loss (ARHL) is highly prevalent among older adults and has been linked to cognitive decline. However, the specific cognitive domains most vulnerable to ARHL and whether these associations exhibit lateralized effects remain unclear, which is critical for understanding and mitigating its broader impact on neurocognitive function. This study aimed to characterize the clinical profile of ARHL and examine associations between hearing thresholds and cognitive performance across domains, including the influence of educational attainment as a proxy for cognitive reserve. Methods: Audiometric assessments and cognitive screening using the Mini-Mental State Examination were conducted in older adults, including normal-hearing listeners (NHL, n = 31, mean age 71.4) and those with hearing loss (HL, n = 46, mean age 73.1). Associations between pure-tone averages, clinical complaints, and cognitive domains were analyzed while considering educational attainment. Results: HL participants exhibited a higher prevalence of tinnitus (NHL: 33.3% vs. HL: 65.2%) and slightly more frequent dizziness compared to their normal-hearing peers. Cognitive assessment revealed that decreased cognitive performance was strongly associated with hearing loss (p < 0.05), and this association was influenced by low educational level. Orientation was the most affected domain (p < 0.01), while recall and language were also significantly associated with low- and high-frequency pure-tone averages, respectively. Conclusions: These findings reinforce the relationship between ARHL and cognitive decline, suggesting an attentional basis whereby higher listening effort to decode the degraded auditory input may affect cognitive performance. The results also highlight the influence of educational attainment as a moderating factor. Full article

In real-world applications, the prevalence of heterogeneous graph data has driven the development of heterogeneous graph neural networks (HGNNs) as an effective solution for modeling intricate semantic relationships. A widely adopted strategy involves using meta-paths as high-level structural motifs to direct neighborhood aggregation in HGNNs. Nevertheless, the semantic content inherent in meta-paths themselves is often not fully exploited, even though they are typically employed as guiding signals. This paper introduces a new HGNN architecture that utilizes meta-path sequences, integrating the intrinsic information of meta-paths directly into the semantic fusion mechanism. By representing meta-paths as sequential data—similar to sequences in natural language processing—we are able to capture more detailed semantic patterns through the sequential order of node types in heterogeneous graphs. Using sequence modeling methods, our approach embeds meta-path semantics into the graph neural network, offering not only additional structural insights but also enabling the training of specialized embeddings for node types. We perform extensive experiments, comprising comparative and ablation analyses, on a custom-built dataset and three publicly available medium-scale heterogeneous graph benchmarks. The experimental outcomes validate the efficacy of our method in utilizing sequential characteristics of meta-paths to improve representation learning. Full article

Influenza A virus remains a major cause of respiratory disease worldwide and poses a persistent challenge to vaccine development due to its rapid genetic evolution and antigenic variability. T-cell-based immunity has therefore gained increasing importance, as it can provide broader and more durable protection by targeting conserved viral regions. Accurate identification of T-cell epitopes (TCEs) is a fundamental requirement for epitope-based vaccine design and immunological research. Although numerous computational methods have been proposed, many existing approaches rely on handcrafted physicochemical features, which offer limited ability to capture contextual sequence dependencies. In this study, a transformer-based viroinformatics framework is proposed for the binary prediction of TCEs from Influenza A virus peptide sequences. The framework employs a pretrained Evolutionary Scale Modeling-2 (ESM-2) protein language model (PLM) to generate rich, contextualized embeddings directly from raw amino acid sequences, eliminating the need for manual feature engineering. These embeddings are processed using a lightweight attention-based transformer classifier to learn epitope-specific sequence patterns. The model achieves strong and stable predictive performance, attaining an accuracy of approximately 97% and an AUC close to 0.99 under stratified cross-validation. Ablation analysis further confirms that protein language model representations and self-attention contribute substantially to performance gains over classical machine learning baselines. To enhance practical reliability, Monte Carlo dropout is incorporated during inference to provide uncertainty-aware predictions, enabling differentiation between high-confidence and ambiguous peptide candidates. In addition, attention-based interpretability is used to identify residue-level contributions to model decisions, offering biologically meaningful insights into epitope recognition. Overall, this study demonstrates that PLMs combined with Transformer architectures provide an effective, interpretable, and a promising computational framework for Influenza A TCE discovery and vaccine research. Full article

Background: Precision nutrition (PN) aims to personalize dietary guidance by accounting for inter-individual variability across biological, metabolic, lifestyle, and environmental factors influencing nutritional needs and health outcomes. While traditional Artificial Intelligence (AI) has advanced nutritional research through systems like automated dietary assessment, these models often operate rigidly. Generative AI (GenAI) introduces the capacity for adaptive interventions for enhanced PN. However, the scope and maturity of its applications remain insufficiently characterized. Objective: This review examined original works applying GenAI in PN, focusing on application, methodology, and limitations. Methods: A systematic search was conducted in PubMed, ACM Digital Library, and Scopus. Inclusion criteria focused on original works deploying GenAI models in PN contexts. Included works were further formally assessed based on data used, validation, transparency, bias, and security and privacy. Results: 21 eligible studies were identified, all published after 2024. The literature indicated a surge in large language model-based systems for personalized dietary recommendations, followed by applications in data foundation building and food effect understanding. A recurrent limitation was questionable evaluation on synthetic data and hallucinations, necessitating a human-expert-in-the-loop, especially in high-stakes clinical settings. Additionally, only 4 of 21 reviewed studies incorporated biological content or biological inputs, and fewer approached biologically grounded PN within implemented personalization workflows using metabolic and/or genomic variables. Conclusions: Although GenAI research in PN is expanding rapidly, most applications remain personalized at a user-preference level rather than including biological determinants. The need for standardized reporting, stronger genome-informed modeling, and consistent human-in-the-loop validation protocols is further highlighted to advance towards holistic PN. Full article

The lack of high-quality linguistic resources, especially large and diverse Arabic dialect corpora, is a major challenge in the development of Natural Language Processing (NLP) applications. By taking advantage of the generative power of Large Language Models (LLMs), this research proposes an efficient approach for the automatic construction of a large-scale corpus of Saudi dialects. We specifically translated 51,840 sentences from Modern Standard Arabic (MSA) into three major Saudi dialects: Qassim (Central), Makkah/Jeddah (Western), and Al-Ahsa (Eastern) using Google’s Gemini 1.5 Pro model. Only two items were flagged by the system as invalid outputs and removed, yielding a pipeline-level invalid output rate below 0.01%. Both quantitative and qualitative differences between MSA and its dialects were discovered through extensive linguistic analyses. Although dialectal sentences had significantly higher lexical density and type token ratios, they were always shorter and more concise. These results suggest that the generated dialectal outputs reflect expected tendencies of informal registers in this controlled, domain-specific setting, while highlighting persistent challenges for dialectal NLP—particularly orthographic variation and the lack of standardized spelling. Full article

Large language models (LLMs) show promise for grading open-ended assessments but still exhibit inconsistent accuracy, systematic biases, and limited reliability across assignments. To address these concerns, we introduce SURE (Selective Uncertainty-based Re-Evaluation), a human-in-the-loop pipeline that combines repeated LLM prompting, uncertainty-based flagging, and selective human regrading. Three LLMs—gpt-4.1-nano, gpt-5-nano, and the open-source gpt-oss-20b—graded answers of 46 students to 130 open questions and coding exercises across five assignments. Each student answer was scored 20 times to derive majority-voted predictions and self-consistency-based certainty estimates. We simulated human regrading by flagging low-certainty cases and replacing them with scores from four human graders. We used the first assignment as a training set for tuning certainty thresholds and to explore LLM output diversification via sampling parameters, rubric shuffling, varied personas, multilingual prompts, and post hoc ensembles. We then evaluated the effectiveness and efficiency of SURE on the other four assignments using a fixed certainty threshold. Across assignments, fully automated grading with a single prompt resulted in substantial underscoring, and majority-voting based on 20 prompts improved but did not eliminate this bias. Low certainty (i.e., high output diversity) was diagnostic of incorrect LLM scores, enabling targeted human regrading that improved grading accuracy while reducing manual grading time by 40–90%. Aggregating responses from all three LLMs in an ensemble improved certainty-based flagging and most consistently approached human-level accuracy, with 70–90% of the grades students would receive falling inside human-grader ranges. A reanalysis based on outputs from a more diversified LLM ensemble comprised of gpt-5, codestral-25.01, and llama-3.3-70b-instruct replicated these findings but also suggested that large reasoning models such as gpt-5 might eliminate the need for human oversight of LLM grading entirely. These findings demonstrate that self-consistency-based uncertainty estimation and selective human oversight can substantially improve the reliability and efficiency of AI-assisted grading. Full article

This study examines the first corporate disclosures issued under the IFRS Sustainability Standards, with full alignment to IFRS S2, using natural language processing and text mining techniques, and contributes evidence to an underexplored phase of sustainability reporting research. Focusing on an emerging market setting, the analysis covers the 2024 reports of 18 firms included in the Borsa Istanbul Sustainability 25 Index. The reports are evaluated through readability metrics (Flesch–Kincaid, Gunning Fog, and SMOG), conceptual concentration measures (TF–IDF), semantic proximity analysis (Cosine Similarity), and network-based methods. The findings indicate a strong degree of technical discipline and standard adherence in the first year of implementation, alongside a pronounced barrier to linguistic accessibility. Average Gunning Fog and Flesch–Kincaid scores of 18.94 and 14.90 suggest that meaningful interpretation of these disclosures requires advanced academic proficiency. The observed technical density reflects the detailed and standard-driven structure of IFRS-based sustainability reporting and points to a persistent tension between technical precision and interpretability, consistent with the Managerial Obfuscation perspective (H1). High levels of semantic overlap further indicate that, under conditions of reporting uncertainty, firms rely heavily on established disclosure patterns, reinforcing professional convergence through both coercive (regulatory alignment) and mimetic (uncertainty-driven emulation) isomorphism (H2). In contrast, distinct narrative configurations identified through principal component and network analyses are evaluated as potential credibility-enhancing signals within the framework of Signaling Theory (H3). Overall, IFRS Sustainability Standards reporting functions in emerging markets as a learning-oriented and strategically relevant disclosure mechanism that may potentially mitigate information asymmetry through its linguistic properties. Full article

Background: Artificial intelligence (AI) shows promising results in lymphoma detection, prediction, and classification. However, translating these findings into practice requires a rigorous assessment of potential biases, clinical utility, and further validation of research models. Objective: The goal of this study was to summarize existing studies on artificial intelligence models for the histopathological detection of lymphoma. Design: This study adhered to the PRISMA Extension for Scoping Reviews (PRISMA-ScR) guidelines. A systematic search was conducted across three major databases (Scopus, PubMed, Web of Science) for English-language articles and reviews published between 2016 and 2025. Seven precise search queries were applied to identify relevant publications, accounting for variations in study modality, algorithmic architectures, and disease-specific terminology. Results: The search identified 612 records, of which 36 articles met the inclusion criteria. These studies presented 36 AI models, comprising 30 diagnostic and six prognostic applications, with Convolutional Neural Networks (CNNs) being the predominant architecture. Regarding data sources, 83% (30/36) of datasets utilized Hematoxylin and Eosin (H&E)-stained images, while the remainder relied on diverse modalities, including IHC-stained slides, bone marrow smears, and other tissue preparations. Studies predominantly utilized retrospective, private cohorts with sample sizes typically ranging from 50 to 400 patients; only a minority leveraged open-access repositories (e.g., Kaggle, TCGA). The primary application was slide-level multi-class classification, distinguishing between specific lymphoma subtypes and non-neoplastic controls. Beyond diagnosis, a subset of studies explored advanced prognostic tasks, such as predicting chemotherapy response and disease progression (e.g., in CLL), as well as automated biomarker quantification (c-MYC, BCL2, PD-L1). Reported diagnostic performance was generally high, with accuracy ranging from 60% to 100% (clustering around 90%) and AUC values spanning 0.70 to 0.99 (predominantly >0.90). Conclusions: While AI models demonstrate high diagnostic accuracy, their translation into practice is limited by unstandardized protocols, morphological complexity, and the “black box” nature of algorithms. Critical issues regarding data provenance, image noise, and lack of representativeness raise risks of systematic bias, hence the need for rigorous validation in diverse clinical environments. Full article

As exploratory learning (EL) is increasingly observed with the use of large language models (LLMs), students demonstrate notably varied levels of engagement and effectiveness when they interact with such LLM-supported learning environments. However, the underlying mechanisms driving these disparities, particularly in how students interact with LLMs, remain underexplored. To address this gap, this observational comparative study systematically investigates the EL strategies of 46 students in two different regions of Asia, classifying 25 distinct strategies across cognitive stages using the Depth of Knowledge model. Our analysis compares strategy usage between high and low-performing student subgroups. The findings reveal: (1) A declining trend in the utilization of EL strategies across ascending cognitive stages. (2) High AWP students employed EL strategies more frequently than their peers, with ten EL strategies exhibiting significant between-group differences. (3) Among students with different AI experience, only a few EL strategies usage and cognitive stages showed significant differences. These insights can help educators and LLM interface designers develop targeted exploratory learning assistance for different types of students and help them build high-level metacognitive processes for effective human–computer interaction. Full article

Background/Objectives: Large language models (LLMs) are increasingly used for patient education, yet comparative evidence regarding their accuracy, safety, and ethical performance remains limited, particularly in high-risk fields such as anesthesiology. This study aimed to conduct a multidimensional comparison of five contemporary LLMs in answering common patient questions in anesthesiology. Methods: In this cross-sectional, comparative in silico study, 30 standardized patient questions covering general anesthesia, spinal/epidural anesthesia, and peripheral nerve blocks were submitted to ChatGPT, Gemini, Microsoft Copilot, DeepSeek, and Grok. Responses were independently evaluated under full blinding by five senior anesthesiology professors using a 5-point Likert scale across six domains: accuracy, safety, completeness, understandability, ethics, and overall assessment. Inter-rater reliability was assessed using intraclass correlation coefficients (ICC). Performance differences were analyzed using linear mixed-effects models accounting for question- and evaluator-level variability, with results reported as estimated marginal means. Results: Inter-rater agreement was good to excellent across all domains (ICC > 0.75). Significant model-related differences were observed for overall assessment, accuracy, safety, completeness, and ethics (all p < 0.001), whereas understandability did not differ significantly between models. ChatGPT achieved the highest overall performance, while Gemini demonstrated superior accuracy. Model performance varied across anesthesiology subspecialties, with significant model × topic interactions identified in multiple domains (p < 0.01). Conclusions: LLMs may serve as supportive tools for patient education in anesthesiology; however, their performance varies substantially across models and clinical contexts. Differences in accuracy, safety, and ethical performance highlight the need for cautious, context-aware integration of LLMs into clinical practice rather than their use as substitutes for anesthesiologists’ clinical judgment. Full article