共找到 20 条结果
Healthcare systems in regions undergoing structural and demographic transitions face persistent workforce shortages, particularly in nursing. Digital and robotic technologies offer opportunities to relieve staff of routine, non-nursing tasks and support more resilient care processes. The KoRob project - an interdisciplinary initiative in the Lusatia region of Brandenburg and - addresses this challenge by developing a collaborative social-robotic system for automating meal delivery in hospital wards. In its early phase, KoRob focuses on requirement analysis, co-creation with nursing teams to inquire early trust towards robots, and identifying technical, spatial, and organizational constraints for safe deployment. To assess staff expectations prior to hands-on experience, the project applies the Individual-Task-Technology Fit (ITTF) framework, examining how cognitive style, openness to experience, perceived task-technology fit, and trust shape early acceptance of autonomous delivery robots. This socio-technical and informatics-oriented approach outlined in this study protocol is Work in Progress and aims to generate transferable insights for integrating collaborative robotics into everyday clinical workflows in transition regions.
Population ageing has intensified long-term care demands globally, necessitating innovative, data-driven, and technology-enabled approaches to support caregivers. This study aimed to assess mental health and quality of life among caregivers of older adults in the Northeast of Thailand and to examine structural relationships among key determinants. A cross-sectional analytical study was conducted between January and August 2025 among 438 caregivers using multistage stratified random sampling. Standardized instruments were used. Multivariate logistic and linear regression analyses, along with Structural Equation Modeling were conducted to identify predictors and examine direct and indirect pathways. The prevalence of depression was 24.2% (95% CI: 20.2-28.5). High caregiver burden (AOR = 3.48, 95% CI: 2.12-5.71), poor social support (AOR = 3.02, 95% CI: 1.82-5.01), and low income (AOR = 2.31, 95% CI: 1.29-4.14) were significantly associated with depression. Social support positively predicted QoL (β = 0.38, p < 0.001), while depression negatively influenced QoL (β = -0.44, p < 0.001). SEM demonstrated good fit (CFI = 0.95, RMSEA = 0.045), confirming depression as a mediator. Caregiver burden and inadequate social support emerged as key determinants, highlighting opportunities for integrating predictive analytics and digital health interventions in long-term care systems.
According to World Health Organization, regular physical activity is proven to help prevent and manage noncommunicable diseases such as heart disease, hypertension, diabetes and several cancers as well as to promote health during specific physiological or metabolic conditions such as pregnancy and obesity. In this context, therapeutic exercise guidelines have been developed to provide structured recommendations for safe and effective exercise prescription across diverse health conditions. Despite their clinical importance, these guidelines are not effectively integrated into digital health systems, as they are typically represented in a descriptive rather than computational form. Most mobile health applications support generic activity tracking without formally representing protocol-based exercise prescriptions or quantifying adherence beyond simple completion metrics. This paper proposes a modeling framework for translating Frequency-Intensity-Time-Type (FITT)-based therapeutic exercise guidelines into computable protocol structures and measurable adherence indicators. The framework introduces (i) a protocol formalization layer encoding guideline parameters as structured entities, (ii) an adherence modeling mechanism defining compliance metrics, and (iii) an explainability module combining rule-based evaluation with AI-assisted natural language generation to produce interpretable feedback. Moreover, an event-driven reference architecture is presented, forming the basis of a mobile application prototype that operationalizes the modeling framework.
Vibe coding tools enable software generation through natural language prompts powered by generative artificial intelligence, substantially lowering the technical barrier to application development. Their potential to empower health professionals without programming backgrounds is promising, yet remains largely unexplored in the literature. To analyze whether vibe coding tools effectively bridge the gap between non-technical health professionals and digital health application development, and to identify the enablers, barriers, and risks associated with their use. This study used a design science approach to create and preliminarily evaluate digital health artifacts. In a 3-hour in-person workshop, medical students used a vibe coding platform to build mHealth prototypes for dementia care challenges. The process included problem identification, ideation, prototyping, presentation, and reflection. Data came from participant observation, artifact review, and a post-workshop satisfaction survey. All three groups successfully developed functional prototypes within the allotted time. Key enablers included conversational accessibility, immediate visible results, and direct clinical applicability. Critical barriers included unfamiliarity with health data privacy regulations, absence of security measures in all prototypes, and a tendency to define excessively broad problem scopes. Notably, no group implemented user authentication or data encryption. Vibe coding effectively brings health professionals closer to digital application development, but this democratization is not sufficient on its own. It must be accompanied by training in data security, health informatics and AI fundamentals.
Digital Twins (DTs) are gaining attention as a promising approach for next-generation healthcare systems. However, their real-world adoption remains limited, mainly due to fragmented data environments and a lack of effective interoperability across clinical systems. Although much of the current research focuses on modeling and advanced analytics, the role of interoperability as a structural enabler has received less attention. This paper examines interoperability as a core requirement for healthcare DTs. It considers how three commonly used standards, HL7 FHIR, openEHR, and the OHDSI OMOP Common Data Model, support different needs across the DT lifecycle, including data exchange, semantic representation, and data reuse for analysis. Each of these standards is effective in specific contexts. OMOP, for instance, is widely used for cohort studies, openEHR supports structured longitudinal records, and FHIR is effective for system integration and data exchange. However, developing more complete DT solutions requires coordinating these complementary capabilities. The paper takes a lifecycle-and role-based view, suggesting that DTs are better understood as the result of coordinated data infrastructures rather than standalone systems. This lifecycle framework provides actionable guidance for scalable DT architectures.
As care for old adults increasingly shifts to the home, integrating everyday health data into clinical practice remains a major challenge. Current solutions based on passive sensors or wearable data often lack contextual understanding, leaving clinicians disconnected from the lived experience at home. To address this gap, we propose a user-centered, natural language-based, asynchronous platform enabling patients to communicate relevant health events and contextual insights in their own words. Developed through a participatory design process, our web-based prototype integrates a large language model with interfaces visualizing health and sensor data, and facilitates communication with hospital care teams. The system supports real-time data sharing, thereby contributing to bridging the hospital-home divide. Early feedback has informed iterative technical refinements, and a real-world user study is in progress. This approach represents an important step toward empowering older adults as active partners in their care and enabling more individualized, responsive clinical decision-making.
Gender inequities persist in the specialist digital health workforce, influencing education, remuneration, and career progression. Despite women comprising the majority of this workforce, disparities in pay and gender remain underexplored. Data were drawn from the 2023 Specialist Digital Health Workforce Census, which included 425 Australian respondents who completed gender-related questions and reported remuneration, was analysed to examine associations between gender, remuneration level, and perceptions of equity. Women represented 74.1% of respondents, yet significant correlations emerged between gender and remuneration (p=0.006). Women earning less than $3,000 per week were more likely to identify financial barriers to career development and emphasise policies on pay equity as facilitators. Perceptions of organisational support for gender equity differed by gender (p<0.001), while remuneration influenced views on career development and work-life balance. Findings highlight structural inequities shaped by gender and remuneration, underscoring the need for targeted interventions such as mentorship, sponsorship, and inclusive leadership programs. Policies promoting pay equity and culturally responsive career development are critical to advancing gender equity in digital health. Further research is required to explore these significant themes.
Large Language Models (LLMs) and knowledge models (KMs) are increasingly integrated to improve clinical decision support systems. We conducted a scoping review to examine the synergy between LLM- and KM-approaches for knowledge model construction, enrichment, and LLM optimization. The selected studies were categorized into three clusters: knowledge-grounded reasoning, LLM-driven knowledge model engineering, and hybrid approaches. Results indicate that retrieval-augmented generation (RAG) grounded in KMs improve LLM reliability, explainability, and adherence to clinical practice guidelines. Conversely, LLMs demonstrate good performance in automating key tasks for KM development, including named entity recognition, relation extraction, and knowledge fusion. Hybrid frameworks combining both paradigms further improve performance and safety. Overall, the integration of LLMs and KMs yields more reliable, scalable, and interpretable AI systems for healthcare applications.
As healthcare demand continues to rise and resources remain limited, many health services have introduced virtual front door (VFD) models-typically using hotlines and/or web portals-to streamline clinical triage and reduce unnecessary emergency department visits. This study aimed to explore culturally and linguistically diverse (CALD) consumers' experiences and barriers to accessing healthcare (including through the VFD) in New South Wales. Semi-structured interviews with CALD consumers and clinicians revealed key barriers, including low awareness, low confidence, low trust or a lack of reassurance, and mixed expectations of digital experiences. Co-designing healthcare access initiatives such as the VFD with CALD consumers is recommended to improve their experience and uptake, thereby maximising the impact of these initiatives.
This paper examines the quality impact of technology on health care services and attitudes, focusing on telemedicine (TM) and artificial intelligence (AI). The aim is to systematically review in the last five years international literature to analyze the benefits and challenges associated with the implementation of these technologies. Emphasis is given to the role of AI and TM, concerning improving accessibility in healthcare and managing economic impact and health challenges through data and coordination of activities. Following PRISMA guidelines, 25 systematic reviews published between 2021 and 2026 were analyzed using a structured data extraction process. This study aspires to provide valuable insights to healthcare professionals and researchers, contributing to the development of best practices for continuous improvement of healthcare quality.
Hearing loss is a rising health concern. The World Health Organization estimates that over 1.5 billion people worldwide are currently experiencing hearing loss. This estimate is projected to rise to 2.5 billion by 2050. Hearing loss is often linked with other health conditions, such as dementia, exacerbating its impact. Chronic hearing loss often worsens over time. Early intervention and effective management are therefore important. One of the first intelligent agents designed to provide personalised advice to help preserve the hearing of individuals experiencing mild hearing loss is presented. The intelligent agent functions as a smartphone app that uses natural language processing to extract information from end-users with mild hearing loss and provides evidence-based personalised advice generated through generative artificial intelligence. The advice was assessed through a tailored criterion inspired by the two validated instruments DISCERN and PEMAT scores. Assessment was undertaken across three dimensions: Relevance (measuring how well the advice is personalised to the user), Accuracy (measuring whether the provided information is accurate), and Understandability (measuring whether the text is easily understandable without specialised knowledge). Measurement across 83 AI-generated responses resulted in the following scores: Relevance (85.5%), Accuracy (80.7%) and Understandability (89.2%).
This study updated an existing health information technology maturity and staging model for nursing homes to support alignment with value-based care. Using a five-round Delphi process (November 2023-July 2024), 22 national long-term care health information technology experts reviewed and refined a validated model of 183 items across 27 content areas. Consensus-based revisions resulted in a streamlined model with 142 items across 21 content areas organised into four domains: resident care, nursing care, clinical support, and administrative activities. Most items (86%) were assigned to Stage 4 or higher, indicating readiness for bi-directional data exchange and resident-centred data use. The revised model provides an expert-informed, policy-relevant framework to guide nursing home health information technology development and value-based care readiness by 2030.
Cardiovascular diseases are the leading global cause of mortality. Artificial Intelligence (AI) offers a potential solution to diagnostic challenges, especially in resource-limited settings. In Morocco, AI integration offers a promising solution to bridge healthcare disparities and expand specialized cardiac care to remote regions. A systematic review was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 guidelines using PubMed, Scopus, and Web of Science. Twenty studies were analyzed, comparing AI-based approaches to standard care in triage, electrocardiography (ECG), imaging, and therapeutics. AI improved diagnostic performance, enhanced detection of silent conditions, and reduced inter-observer variability in ECG and imaging. It also showed potential in optimizing therapeutic decisions. However, generalizability is limited by methodological heterogeneity, a lack of multi-center trials, and insufficient data on hard clinical outcomes like mortality. Routine implementation depends on future prospective studies, more representative datasets, upgraded digital infrastructure, and established ethical and regulatory frameworks.
Large language models are increasingly used for health information delivery, but their value for geographically grounded digital health information services remains unclear. We developed a domain-specific, LLaMA-based Digital Health Index (DHI) chatbot grounded in digital health datasets and compared it with GPT-5.3 in a controlled benchmark evaluation. The benchmark included 15 items spanning five scenario types and three difficulty levels. Responses were independently rated by five health informatics experts using a multidimensional 5-point Likert rubric. Overall, the DHI chatbot outperformed GPT-5.3 across all six core domains, with the largest gains in geographic handling, evidence transparency, and accuracy. These advantages remained statistically significant after false discovery rate correction and were broadly consistent across difficulty levels. The findings suggest that geographically grounded, domain-specific conversational AI may better support accurate, transparent, and locally relevant digital health information services than general-purpose models.
Older adults are particularly vulnerable to the effects of heatwaves and extreme heat due to physiological changes often exacerbated by chronic diseases and decreased mobility. Although cool indoor temperatures can moderate the health consequences of extreme outdoor heat to some extent, it is important to recognize that access to cool indoor temperatures is distributed unequally across population groups. This study aimed to assess the mediating effect of (indoor) domestic heat exposure on the relationship between socioeconomic position and heat exhaustion in older adults. In August 2024, 10,000 participants from the DigiHero cohort were invited to take part in an online survey. Of the 5,026 respondents, persons aged 65 or older were included in the analysis (n = 1,457). Domestic heat exposure was assessed through self-reported indoor temperatures during the day and at night, as well as the availability of cooler rooms at night. Heat exhaustion was measured using self-reported symptoms. Income and education were used as indicators for socioeconomic position. We applied structural equation modelling to assess the total impact of socioeconomic position (education/income) on heat exhaustion and the portion of this effect mediated through domestic heat exposure. Additionally, the analyses were stratified by sex/gender and degree of urbanization. In the study sample, 54% were female; mean age was 72.8 (SD = 5.1); 50% had a high education level; and 70% were not exposed to domestic heat. Lower income and education were linked to more frequent symptoms of heat exhaustion. Lower income, but not education, was associated with higher domestic heat exposure which mediated the income-heat exhaustion relationship (total effect β = -0.09 (-0.14, -0.03); indirect effect β = -0.01 (-0.02, -0.003)). Sex/gender-stratified analyses showed these associations only among women. Sensitivity analyses indicated a stronger mediation in urban settings. The analysis revealed that lower income result in more frequent symptoms of heat exhaustion in older women, partially due to differential domestic heat exposure. Potential sex/gender differences in the effects of heat exposure warrant further exploration.
Understanding how video gaming relates to psychological well-being remains challenging, as prior studies rely on cross-sectional designs prone to construct overlap and limited longitudinal sampling. This study develops a player-agnostic machine learning framework to predict short-horizon well-being from preceding gameplay behavior. Using the PowerWash Simulator Open Dataset (8,372 players), 34 features are extracted from 2-hour gameplay windows preceding each response. A tertile-based binary classification (low vs high well-being) is constructed, excluding the middle group, with thresholds computed within each training fold to prevent leakage. Five-fold GroupKFold validation ensures zero player overlaps. SHAP is used for interpretability. LightGBM achieved 81.1% accuracy (AUC=89.9%) without prior well-being features. SHAP identified enjoyment, focus, and autonomy as dominant predictors. Short-term gameplay patterns demonstrate strong predictive power for well-being, enabling interpretable, real-time monitoring systems grounded in behavioral and psychological indicators.
Psychosocial support is becoming increasingly important for both patients and caregivers during their illness recovery journey. While various self-care strategies can be recommended by healthcare professionals (HCPs) like nurses, such as exercise and relaxation techniques, systematic guidance for evaluating the therapeutic potential of streaming media (SM) is still lacking. Patients engage with entertainment media on streaming platforms daily. Among these, Korean Wave (K-wave) content, including Korean pop (K-pop) and drama (K-drama), has gained global reach in recent years. Music therapy and narrative therapy have established evidence bases as non-pharmacological therapeutic approaches, but their potential in nursing practice is still unexplored. Through a mixed-method analysis comprising a systematic review of 90 music therapy and narrative therapy studies and a thematic analysis of Netflix's animated movie "KPop Demon Hunters" (KPDH), this study identified nine therapeutic themes that can be used for psychosocial support in nursing practices. These therapeutic themes were compiled together with evidence-based mechanisms, KPDH examples, and observable elements in SM content into KEMA (K-Wave Element Mapping and Assessment) - a digital health humanities tool that nurses can use to facilitate informed discussions through SM content that patients watch and listen to. KEMA's platform-agnostic design and cultural flexibility allow it to be used across a broad range of SM platforms and media genres to advance digital health humanities into evidence-based, actionable clinical practice.
The rapid expansion of digital health has created demand for a specialist workforce capable of leading complex change, yet little is known about how leadership and career stage are distributed across digital health occupations. This study analysed data from the 2023 Specialist Digital Health Workforce Census to examine patterns of career stage and leadership by occupation and gender. Career stage was defined by the time spent within the specialist digital health workforce. A total of 699 valid responses were included in the analysis. Leadership was significantly associated with career stage, with leaders more likely to be early in their specialist digital health career. Career stage and leadership representation varied significantly across occupations. Gender distributions were also significant. These findings highlight the inequities in leadership pathways and the need for targeted workforce planning and leadership development strategies to support an inclusive and sustainable digital health workforce.
Addressing neuropsychiatric symptoms in Alzheimer's Disease and Related Dementias (ADRD) requires scalable, non-pharmacological interventions. This study evaluated the user experience and clinical efficacy of a humanoid Socially Assistive Robot (SAR) in an ADRD care setting. A randomized, 8-week, baseline-controlled intervention was conducted (N=19), utilizing each participant's pre-intervention status as the control. Outcomes included acute mood change, sustained depressive symptoms via the Geriatric Depression Scale (GDS), and user acceptance via the User Experience Questionnaire (UEQ). The SAR achieved "Excellent" UEQ ratings (x > 2.0) across all dimensions, indicating high accessibility and engagement. Analysis revealed a significant acute positive mood boost (p < 0.001) and a sustained reduction in GDS scores over 8 weeks compared to baseline. A strong correlation existed between robotic stimulation and clinical GDS improvement (p < 0.01). Humanoid SARs are highly accepted and efficacious adjunctive tools for improving emotional health in ADRD populations, supporting their integration into formal dementia care protocols.
This research aimed to develop and implement a digital innovation model integrating big data for chronic disease management and monitoring, and to promote digital transformation in community-based elderly care. A mixed-method approach centered on big data technology and public health perspectives was adopted. In pilot communities (n=3), the dynamic demand assessment model using Gradient Boosting Decision Tree (GBDT) achieved significantly higher prediction accuracy than traditional experiential judgment (85.3% vs. 62.4%, p<0.01). In Shandong cohort (n=320), real-time monitoring of blood pressure and blood glucose via smart devices increased hypertension management rate from 54% to 72% (p<0.05) and reduced diabetes complication incidence by 19%. In Beijing cohort (n=150), IoT-based fall monitoring shortened average response time from 22 minutes to 8 minutes and improved treatment success rate by 35%. The model demonstrated favorable data security and technical applicability, but faced challenges of digital divide-only 42% of adults aged 65+ could independently use service-booking apps, with 58% relying on family or volunteer assistance. The big data-driven prediction model is feasible for community elderly care and can facilitate its digital transformation, whereas further simplification of operational complexity is required.