Healthcare Tools Research Articles

Insights from high and low clinical users of telemedicine: a mixed-methods study of clinician workflows, sentiments, and user experiences.

This research paper explores the current state of user experience and digital system design in Uzbekistan, focusing on the challenges users face when interacting with public and private digital platforms. With the growing adoption of e-government services, mobile applications, and digital healthcare tools, the importance of effective human-computer interaction has become increasingly evident. Drawing on global examples, particularly from Scandinavian countries known for citizen-centric digital solutions, this study highlights how human-computer interaction principles such as usability, accessibility, and user-centered design can significantly enhance the efficiency and acceptance of digital systems in Uzbekistan. The research serves two primary goals: to assess existing user experience limitations in widely used Uzbek digital platforms and to evaluate how incorporating human-computer interaction methods can improve system effectiveness and user satisfaction. The study argues that well-designed interfaces and intuitive system interactions are essential for digital transformation, social inclusion, and the successful modernisation of digital infrastructure in Uzbekistan.

The present contribution aims to analyze and highlight the potential of piezoelectric materials in actuation and sensing duties, obtaining reliable high-precision outcomes in cutting-edge applications including medical interventions. This involves high-precision actuations of robotized procedures, as well as monitoring and controlling various physical phenomena via structural sensing. The characteristics of these applications offer enhanced precision machinery and robotic tools, medical robotic precise interventions, and high-accuracy structural sensing. The paper exposed, analyzed, reviewed and discussed different subjects related to piezoelectric actuators, involving their displacement and positioning strategies, piezoelectric sensors, medical applications of piezoelectric actuators and sensors, including robotic actuation for medical interventions, and structural sensing in the monitoring of wearable healthcare tools. Discussions among others on the advantages and limitations of piezoelectric sensors and actuators in general, as well as future research perspectives in medical involvements, are also presented at the end of the article. The specific features in the illustrated applications reflect crucial behaviors in robotic actuation for medical interventions, structural sensing in the monitoring of healthcare wearable tools, and the control of various structural physical occurrences.

Africa’s healthcare systems face profound structural challenges, including fragmented infrastructure, systemic data vulnerabilities, and unreliable medical supply chains, necessitating innovative, context-specific solutions. Blockchain technology has emerged as a transformative tool in global healthcare, yet research remains disproportionately focused on high-income economies, with less than 5% addressing Africa’s unique socio-technical landscape. This study employs a multi-method approach, combining bibliometric analysis, latent semantic analysis (LSA), and the PRISMA framework, to map blockchain-healthcare research in Africa, revealing critical gaps and proposing B4HC (Blockchain for Healthcare), a novel conceptual model tailored to resource-constrained settings. Our findings highlight blockchain’s potential to enhance data security in electronic health records (EHRs), optimize pharmaceutical supply chains, and empower patient-centric innovations like digital health wallets and decentralized telemedicine platforms. By addressing ethical and equitable adoption, this research challenges Eurocentric technological determinism, integrates social determinants of health, and aligns with decolonization agendas to foster inclusive health ecosystems. We provide policymakers with a roadmap for sustainable blockchain adoption and outline future research directions to bridge theoretical and practical gaps in decentralized health systems.

The Oxford Visual Perception Screen (OxVPS) is a screening tool recently developed for visual perception deficits that occur after a stroke, such as difficulties in recognizing objects, faces, and reading. The OxVPS allows for quick and accessible screening through 10 subtests, including naming pictures and matching shapes. Hong Kong has many stroke survivors, but only the wealthy can afford comprehensive cognitive assessments, highlighting the need for affordable screening tools in public healthcare. This study developed the Hong Kong—Oxford Visual Perception Screening test (HK-OxVPS), a translation and cultural adaptation of the OxVPS. Normative data from the Hong Kong neurologically healthy population was collected, and cut-off scores for each subtest were derived from the distribution of scores from 95 native Cantonese-speaking participants (50–95 years old). Comparison of cut-off scores with the cut-off scores of the UK version of the OxVPS found a general trend for lower scores on the HK-OxVPS, even on non-linguistic and non-culturally relevant subtests (intercept in Delta plot analysis = 5.69). Age, education, and visual acuity were not significant influencers of HK-OxVPS test performance ( p -values 0.96, 0.16, and 0.07, respectively). However, qualitative inspection of patterns in the data of participants who were unable to complete specific subtests suggested a relationship between age and education on subtest completion. Further validity and reliability testing, as well as improvements to increase test completion, may be necessary to ensure suitability for use with Cantonese speaking stroke survivors.

Accurate prediction of patient status is essential for improving clinical outcomes in complex diseases such as non-pulmonary tuberculosis. Traditional models often struggle to capture the nonlinear and heterogeneous nature of clinical data. This study presents a weighted hybrid nonparametric ensemble that combines k-Nearest Neighbors, Kernel Regression, LOESS, and Gaussian Process Regression for patient status prediction. Model weights are determined using inverse Root Mean Squared Error (RMSE) from validation data, optimizing the blend of model contributions. Using a real clinical dataset, the proposed hybrid model achieved the lowest RMSE (0.163), outperforming all individual base models, with k-NN and Kernel Regression receiving the highest weights. Visual and statistical analyses demonstrate that the hybrid model not only improves accuracy but also provides interpretable insights into model contributions, a key requirement for clinical deployment. The methodology employs a rigorous 80/20 train-test split and includes both model evaluation and explainability (SHAP analysis), supporting transparency and trustworthiness. Our findings confirm that weighted hybridization of diverse nonparametric models enhances predictive performance and interpretability for patient status prediction. This framework is adaptable to other clinical contexts and can inform the development of robust, explainable decision support tools in healthcare.

This quantitative research investigates the positive impact of Artificial Intelligence (AI) on enhancing human potential in daily life. Using a descriptive–correlational survey design, data were collected from 300 adult respondents across diverse professions who actively use AI tools in education, work, healthcare, and communication. The study employed a 20-item Likert-scale questionnaire to measure two variables: AI Usage (independent variable) and Human Potential (dependent variable). Data were analyzed using descriptive statistics, correlation, and regression methods. Results show a significant positive correlation (r = 0.64, p < 0.01) between AI adoption and human potential indicators, including productivity, creativity, and life satisfaction. Regression analysis revealed that AI usage accounts for approximately 34% of the variance in overall human potential. The findings confirm that responsible AI integration enhances human capacity and supports innovation, learning, and well-being in modern society.

This study aims to evaluate how the combined use of Industry 4.0 technologies with lean healthcare tools can improve healthcare organizations' sustainability. Sustainability will be assessed from the Triple Bottom Line (TBL), which includes three perspectives: economic, social and environmental. A systematic literature review (RSL) was performed and, after identifying 987 studies and applying the selection criteria, 43 articles published between 2011 and 2022 were analyzed, exploring the combination of I4.0 technologies with lean tools. The most prominent combination identified in the literature was the use of simulation technologies integrated with the value stream mapping (VSM) tool, a core element of the lean methodology. This pairing was primarily applied to enhance the service level indicator. The findings suggest that such combinations are particularly effective in improving efficiency, resilience and internal processes in healthcare organizations. These insights are especially relevant for the development and adaptation of I4.0 technologies to the healthcare context, offering strategic value during periods of instability and uncertainty, such as those experienced during the COVID-19 pandemic. The study presents several correlations worked by the articles, however, when there are two different combinations of technologies and tools for a single TBL indicator, it is not possible to measure the strength of these relationships and, therefore, to infer which one contributes most to performance improvement. This represents the main limitation of the study. Furthermore, a limitation of the study is the exclusion of 13 articles due to the unavailability of full-text access, even through the Portal CAPES (a Brazilian platform with more than 455 research bases). The results of this research can guide hospital managers in identifying combinations of I4.0 technologies and lean tools with the greatest potential to contribute to improving business sustainability, as measured by the TBL. Thus, through this combined use of technologies and tools, health organizations are expected to achieve better performance, offering high quality services to society. The TBL encourages companies and organizations to take responsibility not only for financial profit but also for the social and environmental impact of their activities. This aspect promotes equity, diversity, employee safety, and engagement with the local community and other stakeholders. It is necessary to highlight that the combination of I4.0 technologies with lean tools and the TBL has the ability to mitigate process waste in the healthcare sector. A similar study was found in the manufacturing area and another in the healthcare area focused on evaluating only the simulation technology. Thus, the originality of the research focuses on evaluating the contribution of the combined use of I4.0 technologies (in addition to simulation) with lean tools for business sustainability, measured through TBL, specifically in the health context.

General-domain large language models (LLMs) have emerged as valuable tools in healthcare, however, their ability to understand and perform tasks based on data stored in tabular form has not been explored in Ophthalmology. We aimed to assess OpenAI's Generative Pre-trained Transformer 4o (GPT-4o) performance within real emergency department (ED) eye-related encounters extracted from electronic medical records in tabular format. We input the excel spreadsheet containing the data on 1,419 unique eye-related ED encounters, divided into (1) chief complaint (CC), history of present illness (HPI), and eye examination; (2) CC and eye examination; (3) eye examination only, into GPT-4o via Microsoft's Azure OpenAI Service using chain-of-thought (CoT) prompting and evaluated the diagnosis and assessment performance of the LLM on the presented data. GPT-4o answers were reviewed by board-certified ophthalmologists and classified as (1) GPT-4o provided a correct diagnosis and assessment; (2) GPT-4o provided an incorrect diagnosis and assessment; (3) GPT-4o unable to provide a correct diagnosis as the encounter documentation was incorrect; (4) GPT-4o unable to provide a correct diagnosis as it required ancillary tests. A sample of encounters were reviewed by a second board-certified ophthalmologist for inter-grader agreement assessment. Average accuracy rates were used to evaluate performance and compare statistical significance across scenarios. A second CoT prompting was performed after providing the LLM with the final encounter diagnosis to evaluate disagreement/inconsistencies between the presented documentation and the reported diagnosis. GPT-4o (CoT) overall accuracy was 0.76 (95% confidence interval [CI], 0.74-0.79); no significant difference was found in accuracy when GPT-4o was presented with CC, HPI and eye findings vs. CC and eye findings vs. eye findings only (P = 0.675). The inter-grader agreement kappa was 0.841 (P < 0.001). GPT-4o identified that 6.6% of all encounters did not have EMR documentation that supported the final encounter diagnosis. When encounters with incorrect EMR documentation and encounters with requirements for ancillary tests (5.2%) were excluded, GPT-4o accuracy was 0.87 (95% CI, 0.85-0.89). GPT-4o could accurately synthesize tabular data and provide assessments and diagnoses in real-world ophthalmology encounters, in addition to identify encounters with documentation that did not support the final ED encounter diagnosis. This capability has the potential to support the clinician's diagnosis.

Telemedicine has emerged as a promising tool in obstetric and reproductive healthcare, offering new possibilities for patient-centered care delivery. This literature review explores its impact across key areas, including abortion, assisted reproduction, childbirth, contraception, gestational diabetes, mental health, opioid and smoking cessation, and perinatal care during the COVID-19 pandemic. A structured narrative approach was applied, with studies identified through PubMed and Scopus databases for screening, with selection based on predefined inclusion and exclusion criteria, and synthesized narratively with attention to clinical outcomes, access, satisfaction, and barriers to implementation. Perspectives on the acceptance of telemedicine among healthcare providers, technological advancements enhancing reproductive outcomes, and telemedicine’s pivotal role in maintaining continuity of care during crises, such as the COVID-19 pandemic, are examined. The review also addresses challenges and barriers, including technological proficiency and patient acceptance, while emphasizing telemedicine’s potential to improve accessibility, patient satisfaction, and healthcare outcomes across diverse reproductive health services.

Modern clinical guidelines (CGs), serving as the foundation of evidence-based medicine, predominantly exist in text document formats (PDF, DOC). It makes them difficult to automatically process and integrate into Medical Information Systems (MIS) and Clinical Decision Support Systems (CDSS). Physicians are forced to manually search for, analyze, and apply these recommendations, which is time-consuming process and increases the risk of missing important details. To solve this problem, the authors present a practical methodology for converting text-based CGs into machine-readable clinical guidelines through the application of ontological modeling. The key idea consists of a two-level model for their representation. The external level (for physicians) is a hierarchically structured text, familiar and convenient for reading and analysis. Key elements of this structure are “data containers,” which clearly describe observations, interventions, and the conditions for their application. The internal level (for computer systems) is a formalized knowledge graph into which the content of the “containers” is transformed. This graph, built upon medical ontologies and classifiers, can be automatically processed by a CDSS to generate personalized prompts directly during a physician's work with the Electronic Health Record (EHR).The proposed approach, based on ontological modeling, allows for:Firstly, integrating CGs into the physician's workflow (the CDSS can automatically analyze patient data and suggest relevant recommendations).Secondly, enhancing treatment personalization through the automatic analysis of multiple individual patient parameters during decision-making.Thirdly, facilitating navigation through CGs, as the structured format simplifies the search for needed information and understanding of the relationships between different recommendations. Fourthly, ensuring knowledge relevance (the process of updating machinereadable CGs when new guideline versions appear can be largely automated).The proposed methodology has been successfully tested on relevant CGs in cardiology, and a CDSS prototype was implemented on the IACPaaS cloud platform. Converting CGs into a machine-readable format is a strategic step from a digital document archive to intelligent assistants that save physician time, reduce error rates, and promote strict adherence to the principles of evidence-based medicine at each patient's bedside.Despite their importance, modern clinical guidelines do not contribute to the automation of clinical activities. They are presented in text formats, such as PDF and DOC, which limits their use in digital healthcare. This lecture presents a methodology for creating machinereadable clinical guidelines (MCG) to integrate them into medical decision support systems and medical information systems. The authors propose a two-level ontological model that includes an external-level ontology, which is a representation of MCGs in the form of hierarchically templated texts for doctors, and an internal-level ontology, which is a formalized knowledge graph for machine processing. The authors use a hybrid approach to create MCGs, combining the creation of structured MCGs by specialists with the use of large language models for formalization.

Artificial intelligence (AI) is increasingly integrated into healthcare, changing processes and structures, and thus the practice of healthcare professionals and potentially the role of patients and the healthcare professional-patient relationship. Beyond high-precision AI algorithms, knowledge of how to evaluate and use AI-based results in everyday healthcare is crucial for high-quality and safe care, and a prerequisite for trust. Therefore, this qualitative study aims to explore 1) the general perception of trust in AI used in healthcare and specifically in wound care, 2) the prerequisites for building trust in AI, and 3) the impact of AI on treatment and healthcare professional-patient relationship, all from the perspective of healthcare professionals and patients. Interviews were conducted in 2022/2023 with healthcare professionals specializing in wound care (N = 12) and in 2023 with patients with chronic wounds (N = 10). The interview guide included questions about digitalization in general and AI in particular, as well as trust and the healthcare professional-patient relationship. Our data revealed a limited understanding of AI principles and evaluation of AI-generated outcomes in both groups. Healthcare professionals recognized the potential of AI to provide data-driven suggestions for diagnosis and therapy, acting as a supportive "second opinion". Patients, on the contrary, expressed a preference for their physicians to incorporate AI-generated results into their care, thereby placing their trust in the physician's ability to apply them correctly. Neither group expected significant changes in the healthcare professional-patient relationship. Trust in AI was linked to general trust in digitalization, and healthcare professionals showed greater trust in AI results that were aligned with their existing expertise and were transparently explained. These findings suggest that AI can be a valuable tool for high-quality healthcare, but in-formed use requires meeting key prerequisites, including Explainable AI (XAI) principles and ongoing training.

BackgroundArtificial intelligence (AI) is increasingly being implemented into healthcare: however, successful implementation depends on public acceptance and trust. To foster responsible adoption, it is essential to understand how the public perceives AI in healthcare and to address their concerns with clear and relevant information. This study investigates perceptions of the public to determine predictors of AI acceptance in healthcare.MethodsA cross-sectional online survey was conducted with 1,205 adults in the Netherlands to assess their acceptance of AI in healthcare. Participants were asked questions about general AI attitudes and literacy. Then were randomly assigned to a real-world healthcare AI scenario involving applications in newborn screening, preventive interventions through wearables, chatbots supporting doctor-patient communication, or augmentation of data for AI training in imaging. Using the Unified Theory of Acceptance, we modeled predictors of acceptance, focusing on perceptions of AI effectiveness, enjoyment, concerns, and mood over the scenarios.FindingsAcceptance was strongly and positively correlated with perceived effectiveness (r(1203) = .75, p < .001), enjoyment (r(1203) = .62, p < .001), and attitudes toward the specific AI use (r(797) = .80, p < .001). Regression analysis showed that the model significantly predicted acceptance, F(9, 789) = 246.70, p < .001, explaining 73.5% of the variance. Significant predictors were attitudes towards the AI use in the scenario, perceived effectiveness, enjoyment and positive and negative mood (ps < .05).ConclusionsPublic acceptance of AI in healthcare is driven by the perceptions of its effectiveness, enjoyment and the public mood. In developing communication strategies and AI design it is important to address emotional reactions and highlight AI's healthcare benefits. Enhancing AI literacy and targeting both emotional and cognitive factors are key to successful AI integration in healthcare.Key messages• Emotional responses towards AI, its perceived effectiveness and its enjoyment are key drivers of AI acceptance in healthcare.• Improving AI acceptance through effective communication, emotional engagement, and AI literacy can enhance the successful integration of AI tools in healthcare.

Introduction: The integration of rapidly evolving digital tools and technologies in healthcare has revolutionised patient care by streamlining processes, improving efficiency and enhancing decision-making. However, their adoption in healthcare largely depends on the personal beliefs and experiences of Healthcare Professionals (HCPs). Aim: To explore the perceptions and beliefs of HCPs regarding the use of Digital Healthcare Tools (DHTs) in the Indian context. Materials and Methods: A qualitative exploratory study was conducted from June to August 2024 in various healthcare set-ups in Delhi-NCR, India. Data were collected using semistructured interviews with HCPs from different backgrounds. Participants were selected based on their proficiency in English and Hindi and their interest in digital tools and technologies. The responses were systematically analysed using a descriptive-interpretive thematic approach, mapped to the frameworks of the Technology Acceptance Model (TAM) and the Work System Model (WSM). Results: Of the 25 study participants, 64% (N=16) were female, with a mean±SD age of 35.9±7.5 years and an average professional experience of 10.5±3.4 years. The majority were engaged in direct patient care (doctors and physiotherapists, 36% each; total 72%) and used digital tools for less than 4 hours per day (80%). Digital tools were most frequently used for email and communication (100%), patient engagement and feedback (92%), and financial management (84%). They were less commonly used for teleconferencing (48%), education (40%), radiology and laboratory services (36%), and roster management (12%). Thematic analysis revealed six broad themes: impact and benefits, data security and privacy concerns, technical issues, training requirements, financial concerns, and future expectations, each with 3 to 7 subthemes. While all participants supported the transition to digital tools and digitalisation in Indian healthcare, they also expressed reservations about data breaches, privacy and security, technical skills, infrastructure and training needs, and data sharing. Conclusion: Digital tools offer significant benefits for enhancing the quality of healthcare services. Despite technical and security challenges, the future of healthcare lies in the advancement of digital technologies. Continuous investment in technology and training is essential to harness the full potential of digital tools.

Artificial Intelligence (AI) technologies demonstrate potential as supplementary tools in healthcare, particularly in surgery, where they assist with preoperative planning, intraoperative decisions, and postoperative monitoring. In oral and maxillofacial surgery, integrating AI poses unique opportunities and challenges due to its complex anatomical and functional demands. This study compares the performance of two AI language models, DeepSeek-R1 and ChatGPT-4o, in addressing technical and patient-related inquiries in oral and maxillofacial surgery. A dataset of 120 questions, including 60 technical and 60 patient-related queries, was developed based on prior studies. These questions covered impacted teeth, dental implants, temporomandibular joint disorders, and orthognathic surgery. Responses from DeepSeek-R1 and ChatGPT-4o were randomized and evaluated using the Modified Global Quality Scale (GQS). Statistical analysis was conducted using non-parametric tests, such as the Wilcoxon Signed-Rank Test and Kruskal-Wallis H Test, with a significance threshold of p = 0.05. The mean GQS score for DeepSeek-R1 was 4.53 ± 0.95, compared to ChatGPT-4o's mean score of 4.39 ± 1.14. DeepSeek-R1 achieved a mean GQS of 4.87 in patient-related inquiries, such as orthognathic surgery and dental implants, compared to 4.73 for ChatGPT-4o. In contrast, ChatGPT-4o received higher average scores in technical questions related to temporomandibular joint disorders. Across all 120 questions, the two models had no statistically significant difference in performance (p = 0.270). In comparisons with previous models, ScholarGPT demonstrated higher performance than the other models. While this performance difference was not statistically significant compared to DeepSeek-R1 (P = 0.121), it was statistically significantly higher compared to ChatGPT-4o and ChatGPT-3.5 (P = 0.027 and P < 0.001, respectively). DeepSeek-R1 and ChatGPT-4o provide comparable performance in addressing patient and technical inquiries in oral and maxillofacial surgery, with small variations depending on the question category. Although statistical differences were not significant, incremental improvements in AI models' response quality were observed. Future research should focus on enhancing their reliability and applicability in clinical settings.

As populations age globally, ensuring equitable healthcare access and social support for older adults in underserved urban areas has become increasingly critical. Elderly residents in low-income districts face challenges, including poor living conditions, social isolation, and healthcare access barriers. Community health workers (CHWs) are vital in bridging these gaps, yet their effectiveness is often Limited by resources and training. This qualitative study explores the healthcare and social needs of elderly residents receiving community services, identifies gaps in support systems, and examines the challenges faced by CHWs in delivering care in an underserved urban district. The study took place in Sham Shui Po, a district in Hong Kong with a high concentration of economically disadvantaged elderly residents. Data were collected through 17 semi-structured interviews with older residents and non-governmental organization (NGO) staff, alongside three focus group discussions with CHWs, and were analyzed using thematic analysis. The results showed that senior residents faced poor living conditions, chronic illnesses, and mobility issues, exacerbated by financial constraints and limited healthcare access. Long wait times, transportation challenges, and language barriers hindered medical service use. Many struggled with digital healthcare tools, limiting their ability to manage health independently. CHWs provided vital support but encountered physical strain, inadequate training, and logistical difficulties, highlighting the need for structured training and better resources. Addressing elderly care challenges requires integrated healthcare models, expanded financial and digital literacy programs, and enhanced CHW training and support. Strengthening these areas can improve health outcomes and well-being for aging populations in low-income urban settings.

Digital health tools have the potential to significantly improve the delivery of healthcare services. However, their adoption remains comparatively limited due, in part, to challenges surrounding usability and trust. Large Language Models (LLMs) have emerged as general-purpose models with the ability to process complex information and produce human-quality text, presenting a wealth of potential applications in healthcare. Directly applying LLMs in clinical settings is not straightforward, however, as LLMs are susceptible to providing inconsistent or nonsensical answers. We demonstrate how LLM-based systems, with LLMs acting as agents, can utilize external tools and provide a novel interface between clinicians and digital technologies. This enhances the utility and practical impact of digital healthcare tools and AI models while addressing current issues with using LLMs in clinical settings, such as hallucinations. We illustrate LLM-based interfaces with examples of cardiovascular disease and stroke risk prediction, quantitatively assessing their performance and highlighting the benefit compared to traditional interfaces for digital tools.

Large Language Models are emerging as powerful tools in healthcare, particularly for complex, domain-specific tasks. This study describes the development and evaluation of PEri-operative AI CHatbot (PEACH). It was developed by embedding 35 institutional peri-operative protocols into a secure large language model environment, with iterative prompt engineering and internal testing to ensure clinical relevance and accuracy. The system was tested with a silent deployment using real-world data. Accuracy, safety and usability were assessed. Accuracy was evaluated by comparing the responses from PEACH against institutional guidelines and expert consensus. Deviations and hallucinations were categorised based on potential harm, and user feedback was evaluated using the Davis' Technology Acceptance Model. Updates to PEACH were made after the initial silent deployment to make minor amendments to one of the protocols. In total, 240 real-world clinical iterations were evaluated. First-generation accuracy was 97.5% (78/80), with an overall accuracy of 96.7% (232/240) across three iterations. In the updated PEACH, accuracy improved to 97.9% (235/240), with a statistically significant difference from the null hypothesis of 95% accuracy (p = 0.018). Hallucinations and deviations were minimal (1/240 and 2/240, respectively). There was high usability, with clinicians noting that PEACH expedited decisions in 95% of cases. The κ statistic for inter-rater reliability for PEACH was 0.772 and 0.893 between three iterations, compared with 0.610 and 0.784 for experienced peri-operative physicians. PEACH is an accurate, adaptable tool that enhances consistency and efficiency in peri-operative decision-making. Future research should explore scalability across specialties and its impact on clinical outcomes.

ABSTRACTBackgroundLarge language models (LLMs) have emerged as potential tools in healthcare. This systematic review evaluates the applications of text-generative conversational LLMs in nephrology, with particular attention to their reported advantages and limitations.MethodsA systematic search was performed in PubMed, Web of Science, Embase and the Cochrane Library in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. Eligible studies assessed LLM applications in nephrology. PROSPERO registration number CRD42024550169.ResultsOf 1070 records screened, 23 studies met inclusion criteria, addressing four clinical applications in nephrology. In patient education (n = 13), GPT-4 improved the readability of kidney donation information from a 10th to a 4th grade level (9.6 ± 1.9 to 4.30 ± 1.71) and Gemini provided the most accurate answers to chronic kidney disease questions (Global Quality Score 3.46 ± 0.55). Regarding workflow optimization (n = 7), GPT-4 achieved high accuracy (90–94%) in managing continuous renal replacement therapy alarms and improved diagnosis of diabetes insipidus using chain-of-thought and retrieval-augmented prompting. In renal dietary guidance (n = 2), Bard AI led in classifying phosphorus and oxalate content of foods (100% and 84%), while GPT-4 and Bing Chat were most accurate for potassium classification (81%). For laboratory data interpretation (n = 1), Copilot significantly outperformed ChatGPT and Gemini in simulated nephrology datasets (median scores 5/5 compared with 4/5 and 4/5; P < .01). TRIPOD-LLM assessment revealed frequent omissions in data handling, prompting strategies and transparency.ConclusionsWhile LLMs may enhance various aspects of nephrology practice, their widespread adoption remains premature. Input-quality dependence and limited external validation restrict generalizability. Further research is needed to confirm their real-world feasibility and ensure safe clinical integration.

Telehealth is influencing nursing practice in the management of long-term conditions. As digital healthcare tools become embedded within the NHS, nurses play a critical role in delivering remote care, supporting patient autonomy and adapting to evolving models of practice. This article highlights how the introduction of telehealth has enabled nurses to become more autonomous in clinical decision-making and patients to have a more proactive role in managing their health conditions. Challenges identified include technological barriers, social disconnection among patients, staff wellbeing concerns, and the sustainability of digital initiatives. Telehealth can improve access and efficiency, however its success relies on appropriate integration into clinical workflows, and adequate training and support for staff wellbeing. Hybrid care models that balance digital innovation with the human aspects of nursing are needed. This article examines the contributions of both nurses and patients in telehealth initiatives, while addressing the associated ethical and legal considerations.