Search Results (250)

Artificial Intelligence (AI) translation tools hold strong potential for supporting multilingual classrooms. This mixed-methods study investigates lower secondary school students’ use of and experiences with a real-time AI translation tool embedded in an interactive presentation platform developed by Learnlab in Norway. The tool combines a fine-tuned version of ChatGPT aligned with the Norwegian curriculum and Learnlab AI Translator. We analyze survey data and system logs (N = 148) and four focus-group interviews (N = 19) from five lower secondary schools. The results indicate that the AI translation tool is beneficial for students overall, but particularly for multilingual students. Students used the tool not only to understand difficult words and sentences, but also to explore languages and engage with their peers’ texts across languages. They reported feeling more included in classroom activities suggesting that the tool enabled collaborative meaning-making across languages and perspectives. However, some monolingual students mainly perceived the benefits for others and did not recognize the tool’s potential for their own language learning. There were no statistically significant differences in perceived inclusion or usability across genders or grade levels. The results are discussed in relation to language support, multilingual learning, and inclusion and participation, with the aim of contributing to more equitable learning environments. Full article

As digital technology has become increasingly integrated into rural governance and daily life in China, Internet use among rural residents exerts a multifaceted influence on their mental health. A key mechanism lies in its restructuring of the neighborhood social environment. Uncovering this mechanism is essential for understanding the theoretical and practical connections between rural social transformation and individual well-being in the digital age. This study applied a binary probit model to data from the 2020 China Family Panel Studies (CFPS) to examine the impact of Internet use on the mental health of rural residents. Mediation analysis was used to examine the role of the neighborhood social environment, and the conditional mixed process method was applied to address potential endogeneity issues. Empirical results demonstrate that access to the Internet, along with the breadth and depth of its use all significantly improve the mental health of rural residents. Internet use promotes mental health by strengthening neighborhood relationship and trust, whereas it also negatively affects mental health by suppressing neighborhood identity. Heterogeneity analyses reveal three key dimensions of variation. (1) By usage type: Activities such as gaming, short-video consumption, and WeChat communication show positive associations with mental health, whereas online shopping and learning exhibit non-significant effects. (2) By user group: The mental health benefits are more pronounced among women, less-educated individuals, and middle-aged to older adults. (3) By region: Positive associations are observed in central and western China, with the most substantial effect in the central region. This study elucidates the mechanism through which Internet use affects mental health: the restructuring of traditional, place-based social capital in rural neighborhoods. These findings offer robust empirical support for policies that integrate digital initiatives with the nurturing of local community bonds to improve rural mental health and foster livable and harmonious villages. Full article

This study examines the integration of generative Artificial Intelligence (AI) tools within a Flipped Classroom model to enhance active learning and feedback processes in an English as a Foreign Language (EFL) teaching program. The participants were 242 pre-service EFL teachers enrolled in upper-level courses at a private university in southern Ecuador. Adopting a mixed-methods, design-based research approach, the study incorporated a diagnostic survey, written reflections, post-intervention survey, and focus groups. These instruments explored students’ prior knowledge, perceptions, and experiences regarding AI-supported learning. Findings showed that AI tools such as ChatGPT, Gemini, and Copilot strengthened students’ linguistic accuracy, writing performance, self-regulation, and understanding of pedagogical concepts. AI-generated feedback complemented teacher feedback by providing immediate and clear guidance, promoting iterative revision and deeper engagement with course content. Participants reported increased autonomy, improved time management, and greater readiness to integrate AI into future teaching practices. The results indicate that AI-supported flipped instruction fosters meaningful learning, enhances feedback quality, and develops both linguistic and pedagogical competencies. Full article

Background: Telemedicine may advance endometriosis care, but few initiatives are integrated in outpatient follow-up. A novel telemedicine approach—tele-patient-reported outcome measures (telePROM)—includes an endometriosis-specific questionnaire and phone and video consultations combined with text messaging (chat) with a multidisciplinary endometriosis team. This study explores how healthcare professionals experience telePROM and its integration in clinical practice. Methods: A qualitative study guided by interpretive description methodology. Data were generated through observations and focus group interviews conducted between January 2023 and March 2024 at a referral centre for endometriosis within a university hospital. A purposive sample of ten healthcare professionals comprising physicians, nurses and a medical secretary participated in the focus group interviews. Inductive analysis was inspired by interpretive description and carried out through an iterative process involving four steps, leading to the development of final themes and interpretation. Results: Three themes were identified from analysis: (1) Balancing Personalised Care With Increased Clinical Complexity; (2) Changing Professional Boundaries in a Digitally Supported Care Model; and (3) System Friction and Flexibility when Integrating TelePROM. Conclusions: Telemedicine improved endometriosis care by supporting patient-initiated and personalised consultations. However, sustainable, effective, and safe integration of telemedicine appears to require clinical experience, interdisciplinary collaboration, and supervision. Text communication (chat) proved to be an important element to ensure collection of additional information to complement patient-reported outcomes and it is essential for patient triage; yet it is rarely described in the literature. Ensuring organisational resilience during the digital transformation of healthcare requires ongoing training of healthcare professionals’ communicative and digital competences and may necessitate restructured technical support, including designated telemedicine experts in clinical practice to eliminate technical disruptions. These initiatives may contribute to and support the future implementation of telemedicine in healthcare. Full article

Real-time visual recognition systems integrated with culturally adaptive reasoning are urgently demanded in globalized culinary scenarios. An agent-oriented framework, Agent-based Gastronomy Recommender Enhanced Engine with YOLO (AGREE-YOLO), is proposed in this study, which integrates an optimized lightweight YOLOv13 detector and vision language model (VLM)-driven agents for cross-cultural seafood recipe recommendation. The improved YOLOv13 is equipped with group shuffle convolution (GSConv) modules and Wise-IoU (WIoU) loss, which is validated on a refined underwater seafood dataset targeting sea cucumbers, sea urchins and scallops. It achieves 91.2% precision and 87.3% recall, with 3.9% and 4.2% increments over the baseline model, and maintains 2.0 ms inference speed. Detection outputs are structured and stored in a MySQL database, and a novel ChatFlow pipeline is constructed in the Dify platform to support natural language database querying. VLM-powered agents retrieve structured data and generate culturally tailored recipes and dish images automatically. Operational validation verifies that the end-to-end pipeline realizes seamless conversion from seafood images to personalized cross-cultural recommendations. This work provides an integrated solution for intelligent, culturally adaptive gastronomy in food informatics. Full article

Background/Objectives: Contrast-enhanced harmonic imaging endoscopic ultrasound (CHI-EUS) is a valuable tool for characterizing solid pancreatic tumors. However, interobserver variability remains a significant limitation in clinical interpretation. Artificial intelligence (AI) may offer objective, reproducible assessments, potentially enhancing diagnostic performance. This study compared the diagnostic accuracy and interobserver agreement of nine physicians with varying CHI-EUS experience levels vs. a dedicated AI system and a general-purpose large language model (ChatGPT) on the same 118 histologically confirmed cases. Methods: We conducted a prospective, multicenter, observer-blinded study involving 118 CHI-EUS video cases of histologically confirmed (EUS-FNB) focal pancreatic masses from three tertiary care centers in Romania. Nine readers were stratified into three groups: trainees (<5 years CHI-EUS experience), intermediates (5–10 years), and experts (>10 years). All readers and two AI models received standardized, anonymized 2 min CHI-EUS video clips. A dedicated AI system used a convolutional neural network (CNN) for lesion segmentation and time–intensity curve (TIC) extraction, followed by a feedforward neural network (FNN) for classification. ChatGPT was separately evaluated on the same videos. Diagnostic metrics (accuracy, sensitivity, specificity, positive predictive value [PPV], negative predictive value [NPV], and AUROC) were calculated. Interobserver agreement was assessed using Fleiss’ and Cohen’s kappa statistics. Results: The dedicated AI system achieved an overall accuracy of 95.8% (sensitivity 96.6%; specificity 94.1%) in diagnosing pancreatic adenocarcinoma. Expert readers had a mean accuracy of 78.8% (sensitivity 86%, specificity 61%, and AUROC 0.74), intermediates 80.8% (sensitivity 83%, specificity 75%, and AUROC 0.84), and trainees had a mean accuracy of 67.2% (sensitivity 70%, specificity 60%, and AUROC 0.67). For the most-likely-diagnosis parameter, interobserver agreement was similar between intermediates (Fleiss’ κ = 0.407) and experts (κ = 0.389), while trainees showed lower agreement (κ = 0.203). ChatGPT correctly classified only 14.1% of PDAC cases. Conclusions: A specialized AI model for CHI-EUS video analysis can achieve expert-level performance and reduce diagnostic variability across experience levels. Integration of dedicated AI systems into CHI-EUS interpretation may enhance accuracy and serve as a valuable decision support tool in clinical and training settings. Full article

Higher education institutions are under increasing pressure to strengthen environmental education (EE) due to critical environmental challenges, while also addressing learner support, engagement, and instructional resource constraints. Recent advances in conversational artificial intelligence (AI), particularly generative AI systems based on large language models such as ChatGPT, enable new forms of human–machine cooperation and provide opportunities for interactive guidelines and individualized feedback. This study evaluates AI-supported EE compared with conventional classroom instruction using a quasi-experimental pre-test/post-test research design. Forty undergraduate students from a Libyan university were recruited and assigned to either the AI-supported EE group (n = 20) or a conventional classroom control group (n = 20). Both groups followed the same EE curriculum over eight weeks. Learning outcomes were assessed across environmental knowledge, attitudes, and environmentally responsible behavior using structured instruments. Paired-samples t-tests indicated statistically significant improvements within the AI-supported group across all outcomes (p < 0.05). However, between-group comparisons did not show statistically significant differences. Analysis controlling for baseline differences indicated a statistically significant group effect for knowledge (p < 0.05), while attitudes and behavior remained non-significant. These findings suggest that AI-supported learning may support EE learning for higher education. Full article

Despite significant advances in AI language models, Modern Standard Arabic (MSA) remains a linguistically complex domain in which apparent fluency often masks deeper grammatical instability. This study investigates recurrent grammatical error patterns in ChatGPT-generated Arabic texts, focusing on how these patterns reflect underlying morpho-syntactic challenges and the constraints of probabilistic language generation. Adopting a qualitative, pattern-oriented analytical framework, the study draws on online focus group discussions with secondary-level Arabic teachers, who served as expert linguistic evaluators. Participants collaboratively examined a set of AI-generated texts to identify and interpret systematic grammatical deviations across five key domains: agreement, inflection and case marking, sentence structure, prepositions and transitivity, and cross-linguistic influence. The findings indicate that grammatical errors in AI-generated Arabic are not random but occur as recurring, structured patterns, particularly in contexts involving long-distance dependencies and morphologically complex constructions. These patterns suggest a reliance on surface-level fluency at the expense of deeper grammatical coherence, reflecting limitations in maintaining consistent morpho-syntactic relationships. This study contributes by identifying and characterizing systematic grammatical patterns in AI-generated MSA as interpreted through expert linguistic judgment, offering a qualitative perspective that complements existing quantitative approaches and advances understanding of how large language models engage with morphologically rich languages. Full article

Background: Artificial intelligence (AI) has shown promise in the interpretation of electrocardiograms (ECGs) using signal-based deep learning models. In parallel, large language models (LLMs) have gained increasing visibility in clinical practice, including exploratory applications in ECG analysis. Whether a general-purpose LLM can meaningfully discriminate cardiovascular disease from athlete ECGs during PPS remains unknown. We aimed to evaluate the diagnostic performance of a general-purpose LLM for this task. Methods: In this multicentre diagnostic accuracy study, we evaluated a commercially available LLM (ChatGPT, version 5) in 2950 competitive athletes undergoing PPS. All athletes underwent resting 12-lead ECG, with second- and third-line investigations performed when clinically indicated. The reference outcome was confirmed cardiovascular disease after full diagnostic work-up (n = 450, 15.3%). For each ECG, the LLM generated a numeric score (0–100) representing the inferred likelihood of underlying disease using a standardized prompt and without task-specific fine-tuning. Discriminative performance was assessed using receiver operating characteristic (ROC) analysis. Misclassification patterns were analysed according to International ECG Criteria. Results: GPT-derived scores demonstrated a marked floor effect, with a median value of 0 (IQR 0–2) in both diseased and non-diseased athletes and substantial overlap between groups. The area under the ROC curve was 0.52 (95% CI 0.49–0.55), indicating performance close to random classification. At the Youden-derived threshold, 79% of athletes with confirmed disease were incorrectly classified as negative. False-negative cases were predominantly characterized by borderline ECG patterns (82%), and a substantial number of red-flag ECG abnormalities were also missed. Conclusions: In this PPS cohort, a general-purpose LLM used in a naïve configuration showed no clinically meaningful ability to discriminate between cardiovascular disease and athlete ECGs. Without task-specific training or domain adaptation, such models should not be used for diagnostic triage in athlete screening. Full article

Background: This study aimed to estimate the prevalence and associated demographic factors of autism spectrum disorder (ASD) in children aged 3 to 6 years in Bahrain, as well as to identify co-occurring developmental disorders. Methodology: The study sample comprised 500 children who attended eight health centers across four governorates (Group A) in Bahrain. A second group (Group B) consisted of all children who completed their diagnosis at the Child and Adolescent Psychiatric Unit for ASD from June 2023 to May 2024 to identify associated developmental disorders (n = 232). Group A mothers were interviewed using the M-CHAT-R. For Group B, we used children’s files, the General Intelligence Scale (Stanford-Binet), the M-CHAT-R, the CARS, Conners’ Form, and the Zarit Burden Interview to assess family burden. Additionally, a file review was conducted to determine the presence of intellectual disability (ID) in Group B cases. The Conner-3 Short Forms Test was administered to all cases aged 6 years or older (n = 64), and family burden was assessed using the Zarit Burden Interview Scale for reachable cases (n = 176). Results: Group A findings indicated an overall screen-based/at risk prevalence of ASD of 2.6%, with a higher prevalence in males (1.6%) compared to females (1%). In Group B, there was a78% prevalence of ID and a 17.2% prevalence of attention-deficit/hyperactivity disorder (ADHD). The study also found that the impact of having a child with ASD on the family varied based on the nature and severity of the disorder or disability, with moderate to severe burden reported at approximately 38%. Conclusions: The prevalence of ASD among young children was notably high, particularly among males. The most common comorbidities were ID followed by ADHD. The family burden associated with ASD was significant, with more than one third reporting moderate to severe burden. These data are essential for informing health education and social service planning. Full article

Many tertiary mathematics departments are seeking to improve equity in their programs; however, they may struggle to translate these goals for equity into action. This longitudinal, qualitative study focuses on a Networked Improvement Community (NIC) within the mathematics department at a public, doctoral degree-granting university located in the Southeast United States. This NIC worked together for two years (Spring 2023 to Spring 2025) to become more reflective practitioners and critically transform the mathematics program at their institution. We used Cultural Historical Activity Theory (CHAT) to examine relationships between objects, tools, and outcomes for the NIC. Data included multiple interviews and journals from eleven (n = 11) participants, and was triangulated with observer field notes of monthly NIC meetings. Thematic analysis revealed three pathways that connected NIC members’ individual and collective goals (objects), NIC activities and resources (tools), and NIC members’ perspectives on teaching and students (outcomes). We found that sometimes objects, mediated by tools, led to aligned outcomes, but not always. Specific tools could lead the NIC to adopt a new and collective object (and outcome). In other cases, the lack of the right tool led to unrealized outcomes or even secondary outcomes within the NIC. Ultimately, the critical transformations that NIC members envisioned were not realized; however, the experience of examining student data and discussing with colleagues shaped their thinking about teaching and students in impactful ways that inform faculty development for institutional change efforts on a broader scale. Our findings highlight the importance of identifying the right tools to support critical transformation, including the value of examining data as a collaborative group. We also extend NIC scholarship by using second-generation CHAT to distinguish objects over time and specify pathway models linking tools to outcomes. Full article

Background/Objectives: This study aimed to compare AI-generated educational material with faculty-authored content in Dental Prostheses Technology, evaluating perceived clarity, accuracy, structure, usefulness, and overall instructional quality across different age and professional groups. Methods: An analytical cross-sectional study was conducted using two versions of the first three chapters of a prosthodontics textbook: the original faculty-authored text and a reformulated version generated by ChatGPT 5.2 (OpenAI). Images were removed and formatting standardized to ensure a text-only comparison. An anonymized online questionnaire based on a five-point Likert scale assessed clarity, accuracy, readability, usefulness and structure. To reduce potential bias, participants were unaware of the authorship of the evaluated materials (human-authored vs. AI-generated). A total of 130 participants independently reviewed both documents. Data were analyzed using Wilcoxon signed-rank, Mann–Whitney U, and Friedman tests. Results: Both materials received favorable evaluations across all dimensions. The AI-generated version demonstrated a statistically significant advantage in clarity (Z = −2.107, p = 0.035; r = 0.19), while no significant differences were observed for structure, accuracy, readability, or usefulness. Generational differences emerged: younger participants valued improved clarity but reported reduced usefulness, mid-career participants showed the greatest improvement in perceived accuracy, and senior professionals reported substantial gains in usefulness and readability. Conclusions: AI-generated educational material demonstrates pedagogical equivalence to faculty-authored content, with clarity representing its principal advantage. Large language models may serve as effective complementary tools in dental education, particularly for restructuring complex content. Full article

Large language models (LLMs) are increasingly used in applied intelligent systems, but mid-sized models still lag on mathematical reasoning, partly because reliable step-level supervision is scarce. Many existing remedies rely on costly human annotation, stronger teacher models, or heavy training pipelines, which limits practical adoption. We propose VGPO-MCTS (Value-Guided Group-wise Policy Optimization over Monte Carlo Tree Search), a search-and-distillation framework that constructs reusable step-level supervision from datasets that provide only problems and final answers. VGPO-MCTS augments a frozen backbone with (i) a lightweight value model that scores candidate reasoning states formed by a reasoning prefix and its candidate next step, and (ii) a policy updated with parameter-efficient adaptation. During search, the value model guides tree expansion and selection, while verified outcomes are propagated backward to correct node utilities. The corrected search trees are then distilled into two complementary datasets: a value regression dataset for value learning and group-wise sibling candidate sets for GRPO-style policy optimization. Experiments on GSM8K and the MATH dataset with ChatGLM3-6B and SciGLM-6B show stable round-wise improvements in final-answer exact match under a lightweight adaptation setting. After three rounds of self-training, the proposed framework improves performance by about 6.3 percentage points on GSM8K and about 3.9 percentage points on MATH across the two backbones. Full article

Distance learners in higher education are often assumed to face limited peer interaction, potentially weakening their sense of belonging. This study examines peer relationships and belonging among students in distance and blended university programs, with attention to the role of virtual reality (VR) within digitally mediated learning environments. Immersive VR teaching is included in the curriculum for distance learning students in the studied programs. Using a mixed-methods design, survey data and open-ended responses were collected from 17 students in Information Studies and Information Service Design. An adapted Classroom Community Scale was supplemented with items addressing the perceived contribution of different communication technologies. Contrary to expectations, fully distance learners did not report weaker agreement with statements reflecting belonging than blended students; on several items, they expressed stronger agreement, particularly regarding perceived peer support and learning opportunities. Results indicate that conventional 2D communication tools, particularly chats and video calls, are central to sustaining peer relationships. VR was not perceived as essential but described by some students as an added value supporting shared experience and group cohesion. Overall, belonging emerges as a socio-technical achievement shaped by communication practices rather than physical proximity. Full article

In Vietnamese primary mathematics education, geometry instruction often emphasizes rote calculation and formula memorization rather than meaningful contextualization, leaving students disconnected from abstract concepts and lacking opportunities to connect learning with cultural identity. This quasi-experimental study investigates how integrating generative AI tools (ChatGPT, DALL·E, Canva) with the culturally grounded Vietnamese folktale Bánh Chưng—Bánh Giầy can support Grade 5 students’ understanding of circle geometry. Employing a mixed-methods design with 30 students divided into experimental (AI + storytelling) and control (traditional instruction) groups, the study measured cognitive and affective learning outcomes through pre/post-tests, a validated 25-item questionnaire, interviews, and classroom observations. Quantitative results revealed significant improvements in the experimental group across all measured dimensions, learning interest, attentional focus, conceptual understanding, mathematics passion, and cultural preservation awareness, with large effect sizes. Qualitative findings confirmed enhanced engagement, multimodal conceptual clarity, and cultural affective resonance. The study demonstrates that low-cost, teacher-mediated generative AI can effectively support learning in resource-constrained primary settings when anchored in local narratives. Implications for ethical AI integration and teacher professional development in Vietnamese contexts are discussed. Full article