Diabetes-associated cognitive impairment (DACI) undermines diabetes self-management and is associated with increased morbidity and mortality. Although mind-body exercise may improve cognition, its benefits are often limited by the lack of timely, precise, individualised instruction and feedback in adults with cognitive impairment. Augmented reality (AR) systems can deliver performance-adaptive, visually anchored cues and auditory prompts in real-world settings, but their cognitive and metabolic benefits in DACI remain unknown. This trial will compare AR-assisted Tai Chi (iTaiChi), instructor-led Tai Chi (TaiChi) and standard care in adults with DACI and explore underlying metabolic and neurobiological mechanisms.

Artificial intelligence (AI), machine learning (ML), and deep learning (DL) are reshaping orthognathic surgery and maxillofacial traumatology by supporting diagnostic imaging and three-dimensional virtual surgical planning (VSP). This study provides a bibliometric mapping of the field in order to characterise its thematic structure, trace its temporal evolution, describe international collaboration patterns, and situate the clinical limitations reported in the literature. A systematic search was conducted in Scopus, PubMed/MEDLINE, and IEEE Xplore from database inception to 24 August 2026, restricted to English-language records. The query combined three blocks: artificial intelligence terminology ("artificial intelligence", "machine learning", "deep learning", "neural network*", "machine intelligence"), craniomaxillofacial terms ("orthognathic", "maxillofacial", "mandibular split", "le fort", "stomatognathic"), and surgical terms ("surgery", "surgical", "planning", "osteotomy"). Records were eligible when they both implemented an AI method and applied it to the craniomaxillofacial or dental domain, including educational and patient-communication applications; records were excluded as editorials, errata, letters and short opinion pieces, conference proceedings volumes, studies without an AI method, medical head-and-neck oncology, three-dimensional facial reconstruction without craniomaxillofacial surgical application, non-clinical computer vision, or topics outside the craniomaxillofacial domain. The search retrieved 1,925 records; after removal of 539 duplicates, 1,386 unique records underwent title and abstract screening, and 854 documents met the eligibility criteria. Bibliometric networks were constructed in VOSviewer (version 1.6.21; Centre for Science and Technology Studies (CWTS), Leiden University, Netherlands). Keyword co-occurrence analysis resolved five thematic clusters, centred respectively on generative AI and language models, orthognathic surgery and cephalometry, dentomaxillary radiological imaging, DL methodology, and computer-assisted surgery with digital dentistry. Temporal overlay analysis showed that keywords relating to large language models and educational applications carry the most recent average publication years, whereas cephalometric and landmark detection terms are comparatively older. Country-level co-authorship analysis identified 35 countries meeting a five-document threshold, organised into six regional clusters, with China and the United States as the principal contributors by output and the United States as the most central connector. These bibliometric observations indicate a field in rapid thematic transition, in which methodological heterogeneity, dataset homogeneity, and limited algorithmic interpretability remain the barriers most frequently reported to clinical translation.

This article describes the protocol used to conduct a systematic mapping study (SMS) of software architecture. The presence of architectural and design instruction in primary studies cannot be determined through search and superficial review alone. A principled quality assessment must be conducted to identify studies that are both relevant and of high quality. LLM assistance offers practical means of performing such assessments in a timely and efficient manner. Therefore, our contribution is threefold: • Guidance for the following SMS steps: tooling setup, database search, deduplication and snowballing, primary study selection, quality assessment, and data extraction. • The rationale and methodology to perform quality assessment on primary studies regarding software architecture. • A method of using large language models (LLMs) together with human evaluators is presented. The method is supported by bespoke open-source software.

Flipped instruction has gained traction in computer-assisted language learning, but evidence of its effects on L2 speaking and the underlying affective and motivational mechanisms remains limited, especially for young adult EFL learners in East Asia.

Previous works that have analyzed electroencephalographic (EEG) speech envelope encoding in hearing-aid users using stimulus reconstruction (SR) approaches have shown that noise reduction (NR) algorithms can improve the neural tracking of speech in noise. Beyond enhancing speech envelope encoding, NR may also restore the perceptibility of acoustic edges that trigger auditory late responses (ALRs), likely promoting more consistent ALR elicitation and thus enhancing brain-to-speech synchronization. To investigate this hypothesis, 26 hearing-impaired listeners were fitted with bilateral hearing aids and participated in an EEG experiment with the instruction to listen to a continuous single-speaker speech stimulus under speech-in-quiet conditions and speech-in-noise conditions with and without NR. Event-related potential analyses of neural responses to acoustic edges in the speech onset envelope revealed that P1 component amplitudes in averaged ALRs decreased significantly for speech-in-noise compared to speech-in-quiet conditions, but partially recovered when NR was activated. These observations correlated significantly with the degree of spectrotemporal consistency across single-trials for the associated ALR time-frequency component. Additionally, a SR analysis for the speech onset envelope produced qualitatively identical results that were highly correlated with those from the spectrotemporal consistency analysis. These findings suggest that hearing aid NR algorithms partly enhance neural speech tracking by restoring neural responses to acoustic edges. In turn, our results may support the future development of hearing aid processing strategies that could facilitate neural tracking of speech based on neural principles.

Multimodal artificial intelligence, albeit showing great potential in computational pathology, remains limited to isolated patch-level interpretation and often fails to analyze gigapixel-scale whole-slide images (WSIs) essential for clinical utility. Here we present SlideChat, a multimodal generative artificial intelligence assistant for whole-slide computational pathology across cancer types. SlideChat integrates patch-level and slide-level pathology encoders with a pretrained large language model. Using 274,233 multimodal instruction samples, SlideChat is trained to learn the associations between WSIs and diagnostic reports and interpret complex queries in clinical practice. Evaluated on 8,836 closed-ended questions, 129 open-ended questions and 3,149 WSI reports from five cohorts spanning 31 cancer types, SlideChat outperformed leading baselines by 19.1% in closed-ended accuracy and by 7.7% in report-generation Metric for Evaluation of Translation with Explicit Ordering score and received the highest expert ratings across five dimensions in open-ended question answering, showing the potential to enhance diagnostic workflows, medical education and clinical decision-making.

Children with autism spectrum disorder (ASD) have significant deficits in communication, emotional recognition, perspective-taking, and social skills. Evidence from existing studies has shown that the social interaction skills of children with ASD are closely related to their theory of mind (ToM) abilities. While prior research has explored the effect of single intervention modes such as computer-assisted instruction (CAI) and the thought-bubble strategy, the efficacy of combining CAI with the thought-bubble strategy remains unclear. This study investigates the effectiveness of this multifaceted approach in fostering ToM development in children with ASD. Using a single-case design, the intervention for two children with ASD involved implementing CAI combined with the thought-bubble strategy to teach ToM skills to children with ASD, including basic beliefs, first-order beliefs and first-order false beliefs. Results indicated that both children showed an increase in their ToM abilities, evidenced by their increased tendency to use "I think/I feel" statements in everyday life, enhanced understanding of false beliefs, and a decrease in negative emotions. The thought-bubble strategy can be embedded in CAI, and the combination of these two methods facilitates the enhancement of the ToM abilities among children with ASD with good maintenance and generalization effects.

Understanding enzyme kinetics is fundamental in biomedical science, pharmacology, and biochemistry, yet students often struggle to interpret kinetic models and relate them to biological function. To support learning in this area, an interactive web-based tool, the Kinetics Analysis App, was developed using the Shiny framework in R. The application enables users to input experimental or example data and compare four classical models of enzyme kinetics: Michaelis-Menten, Lineweaver-Burk, Eadie-Hofstee, and Hanes-Woolf. Real-time visualization of fitted curves and parameter estimates (Vmax and Km) allows learners to explore how model choice, data variability, and substrate concentration influence kinetic interpretation. The tool supports scaffolded, constructivist, and multimedia learning approaches, lowering mathematical barriers and encouraging hypothesis-driven exploration. Designed for integration into laboratory classes, flipped learning, or independent study, the app promotes critical thinking and enhances conceptual understanding of kinetic modeling. Informal student feedback indicates increased confidence and clarity when interpreting enzyme kinetics compared with traditional spreadsheet-based approaches. The Kinetics Analysis App therefore offers a scalable and accessible method to improve quantitative skills and deepen pedagogical engagement with enzyme kinetics.NEW & NOTEWORTHY Students often find enzyme kinetics difficult because mathematical models can appear disconnected from biological meaning. The freely available Kinetics Analysis App was created to help address this challenge. Through interactive visualization of four classical kinetic models, learners are able to explore parameter estimation and model selection in real time. This web-based tool supports active learning, improves confidence in data interpretation, and provides an accessible approach for teaching quantitative concepts in biomedical and pharmacological sciences.