Search Results (12,415)

Background/Objectives: The integration of artificial intelligence (AI) in medical devices has significant potential for transforming healthcare in the European Union (EU). However, AI medical devices are regulated under the Medical Devices Regulation (MDR) and the In Vitro Diagnostic Medical Devices Regulation (IVDR), as well as horizontal legislation such as the AI Act, resulting in a multilayered framework that raises difficulties for all stakeholders. This scoping review aims to identify and classify the challenges associated with the regulation of AI medical devices in the EU and the corresponding recommendations proposed in the academic literature. Methods: This review followed the PRISMA-ScR guidelines. The search was conducted in February 2026 in Web of Science, Scopus, MEDLINE PubMed, and IEEE Xplore. Challenges and recommendations were extracted through an analytical review process and grouped into secondary and primary domains, and a composite indicator was developed to assess the predominance of each domain. Results: Seventy studies published between 2018 and 2025 were included. A total of 261 challenges and 113 recommendations were identified and organized in five primary domains: regulatory framework, data issues, accountability and ethics, development and deployment, and certification. The regulatory framework was the domain most frequently reported, reflecting the overlap between the MDR, the AI Act, and other legislation and the absence of harmonized standards. Transparency was the only secondary domain where recommendations outnumbered challenges. Conclusions: This review structures the fragmented evidence into a single framework, mapping areas most frequently discussed in the literature and identifying where recommendations are still scarce, and it provides a foundation for policymaking and regulatory science on AI medical devices in the EU. Full article

Human Digital Twins (HDTs) extend traditional Digital Twin (DT) concepts by modeling both humans and hospital processes to support smarter and more human-centered healthcare. By integrating Industry 4.0 (I4.0) technologies with the human-centric principles of Industry 5.0 (I5.0), HDTs offer new opportunities to improve hospital operations. This study presents a PRISMA-based systematic literature review to examine the role of HDTs in hospital operations. A total of 329 papers were identified through the initial search, and after the screening process, 22 studies were included for in-depth analysis. The review combines bibliometric analysis to examine publication trends, leading authors, contributing countries, and keyword co-occurrence with a content analysis to identify the main research themes. Three major themes emerged: (1) HDT architectures and data integration, (2) human-centric and governance aspects, including explainable artificial intelligence and privacy, and (3) operational and clinical outcomes, including patient flow, resource utilization, and staff support. Based on these findings, the study proposes a four-layer HDT framework for practical implementation in hospital operations. Although the reviewed studies indicate that HDTs have considerable potential to improve operational efficiency and strengthen human involvement, most existing research remains conceptual or simulation-based. Future research should therefore prioritize real-world implementation and validation while incorporating ethical, explainable, and sustainable design principles. Full article

Background/Objectives: Contemporary cardiovascular critical care is increasingly shaped by advanced technologies, prognostic uncertainty, and time-sensitive decisions regarding mechanical circulatory support, treatment limitation, device management, family communication, and end-of-life care. Although established ethical principles provide essential normative guidance, their integration into routine bedside workflows remains challenging. This review examines the principal ethical tensions arising in the Cardiovascular Intensive Care Unit (CICU) and proposes a practical framework for real-time ethical decision-making. Methods: A narrative review of the contemporary literature was performed using major biomedical databases together with international guidelines, scientific statements, consensus documents, and relevant ethics literature to identify the principal ethical challenges encountered in contemporary cardiovascular intensive care and to develop an implementation-oriented framework for integrating ethical reasoning into routine clinical workflows. Results: Ethical challenges in the CICU frequently arise during transitions between therapeutic escalation, reassessment, limitation, withdrawal, and comfort-focused care. These decisions are influenced not only by clinical prognosis but also by organisational culture, multidisciplinary dynamics, patient preferences, communication, and structural inequities. We propose Operational Ethics as an implementation framework that embeds ethical reflection into routine clinical workflows through continuous reassessment of therapeutic goals, proportionality, prognosis, patient values, and multidisciplinary deliberation. Practical strategies include ethical pause points, time-limited trials, structured communication pathways, anticipatory device discussions, regular goals-of-care reassessment, and institutional support for clinicians experiencing moral distress. Conclusions: Operational Ethics complements established bioethical principles by translating them into practical bedside processes adapted to the temporal and organisational realities of cardiovascular critical care. Its integration into routine CICU practice may support more consistent, equitable, and patient-centred decisions while ensuring that technological possibilities remain aligned with meaningful clinical goals. Full article

Smart cities and smart real estate are increasingly used to enhance urban efficiency, sustainability, and quality of life. Nevertheless, their implementation is hindered by interconnected challenges across concepts, technology, governance, economics, society, ethics, and the environment. Thus, this paper examines how and why interconnected limitations constrain the smartness of smart city and smart real estate initiatives and identifies the governance and implementation responses revealed by cross-case comparison. The analysis of secondary academic sources, policy papers, standards, and industry reports compares Toronto, Songdo, NEOM, and Barcelona. The results identify six categories of limitations: conceptual and strategic; technological and infrastructural; governance and institutional; economic and financial; social and ethical; and environmental. The analysis shows that the challenges facing these initiatives go beyond technology alone and include ambiguous definitions of smartness, poor governance, limited public involvement, privacy and surveillance risks, high implementation and maintenance costs, interoperability issues, and unverified assumptions that technological advancement will deliver sustainability. The comparison of cases highlights that the success of these initiatives depends on effective governance, responsible data management, public trust, financial feasibility, institutional coordination, and a life cycle approach. The study concludes that smartness should be understood as an integrated institutional, economic, social, technological, and environmental capacity rather than as the mere adoption of digital technologies. Full article

This study analyses changes in university students’ self-perceived eco-social competences through active methodologies linked to holistic sustainability and human rights. A multidimensional rubric based on the KNOW–CARE–DO model was applied to assess four eco-social competences: (1) critical contextualisation of knowledge in relation to social, economic, and environmental issues; (2) respect for human rights, diversity, equity, a culture of peace and participation; (3) undertaking actions for the common good and sustainable development; and (4) the application of ethical principles linked to sustainability and human rights. A pre-experimental design incorporating pre-test and post-test measurements was used. The study involved 61 students from universities in Spain, Colombia, and Italy. An open-ended question was included to complement the quantitative data. The results indicate significantly higher self-perceived competence scores at post-test than at pre-test, particularly in the critical contextualisation of knowledge. The rubric demonstrates acceptable-to-high internal consistency, although further psychometric validation is required. Full article

Mobility-as-a-Service (MaaS) has emerged as a promising approach to urban transportation by integrating multiple mobility services into a single digital platform for trip planning, booking, and payment. More recently, Artificial Intelligence (AI) has expanded the capabilities of MaaS, enabling more efficient data processing, predictive analytics, personalized services, and intelligent decision support. This narrative review examines the current state of research on AI-enabled MaaS from both technological and socioeconomic perspectives. The analysis covers five major research areas: data integration and interoperability, predictive systems and demand forecasting, AI-enabled decision support for policy and planning, fairness and ethical AI, and cybersecurity and privacy protection. The findings show that successful implementation depends not only on advances in AI algorithms but also on high-quality interoperable data, effective governance, regulatory support, public trust, and collaboration among stakeholders. The review concludes that the main challenges facing AI-enabled MaaS are no longer primarily technical but organizational, institutional, and social. Future research should focus on trustworthy and explainable AI, privacy-preserving learning, standardized evaluation methods, fairness-aware optimization, resilient cybersecurity, and long-term assessments of MaaS impacts on sustainable urban mobility. Full article

Artificial intelligence (AI) and machine learning (ML) are transforming the landscape of rheumatological and systemic inflammatory disease management, offering unprecedented capacity to integrate complex, multidimensional data for diagnostic support, disease monitoring, and therapeutic decision-making. This comprehensive narrative review, based on a non-systematic literature search of PubMed/MEDLINE and Google Scholar combined with the authors’ clinical expertise, provides a clinically oriented synthesis of current and emerging AI applications across the full spectrum of immune-mediated inflammatory diseases—including rheumatoid arthritis, systemic lupus erythematosus, vasculitis, inflammatory bowel disease, psoriatic arthritis, systemic sclerosis, inflammatory myopathies, and sarcoidosis—with particular attention to applications that have demonstrated or are approaching clinical utility. We discuss deep learning-based image analysis, natural language processing of electronic health records, multi-omic biomarker discovery, and the application of Fourier transform-based signal processing to biological time series as a novel approach to continuous disease monitoring. Fourier transform methods—already foundational in MRI reconstruction, cardiac electrophysiology, and clinical neurophysiology—are here systematically extended to rheumatological and inflammatory disease signals, including accelerometry, electromyography, heart rate variability, and longitudinal biomarker time series. The phenomenon of large language model hallucination—particularly critical in rare inflammatory diseases—is addressed alongside retrieval-augmented generation as a mitigation strategy. We further argue that AI-driven methods do not merely improve the interpretation of clinical data, but fundamentally expand what is observable—with profound epistemological implications for clinical knowledge transmitted through generations of medical tradition. Ethical considerations and future directions toward precision inflammatory disease medicine are outlined. Full article

The increasing integration of AI in education imposes the need to examine the competencies of teachers for its pedagogically meaningful, safe, and responsible application. The goal of this research was to examine the self-assessments of teachers’ AI-related competencies, as well as to determine whether there were statistically significant differences regarding the subject area and years of teaching experience. The research was conducted on a sample of 165 primary and secondary school teachers in the Republic of Serbia. Data were collected using the Serbian-language version of the original Teacher Artificial Intelligence Competence Self-Efficacy (TAICS) Scale. The reliability and factor structure of the instrument were checked using Cronbach’s alpha coefficient, Kaiser–Meyer–Olkin (KMO) indicator, Bartlett’s test of sphericity, and exploratory factor analysis using principal axis factoring with Varimax rotation, while one-way analysis of variance (ANOVA) was used to examine differences. The results showed high internal consistency of the scale (Cronbach’s α = 0.972) and an interpretable three-factor solution explaining 71.202% of the total variance. The following factors were extracted: pedagogical–evaluative AI competencies, ethical–safety AI competencies, and operational–technical AI competencies. No statistically significant differences in pedagogical–evaluative or ethical–safety AI competencies were found according to subject area or years of teaching experience. For operational–technical AI competencies, an unadjusted difference was observed according to teaching experience; however, this finding did not remain statistically significant after the Bonferroni correction for multiple testing. These findings should be interpreted as preliminary and require replication in larger and more representative samples. The findings indicate that the competencies of teachers for artificial intelligence should be viewed as a multidimensional basis for further research, improvement of teaching practice, and planning of professional development in the field of application of AI in education. Full article

Background: Rare diseases collectively affect millions of children worldwide and are a major cause of pediatric morbidity, mortality, and lifelong disability. Although most have a genetic basis, obtaining a timely molecular diagnosis remains challenging because of substantial clinical and genetic heterogeneity. Advances in genomic medicine are transforming rare disease diagnosis and establishing genomics as the center of precision medicine. Methods: This review summarizes current evidence on genomic approaches for pediatric rare diseases, including established and emerging sequencing technologies, their clinical applications, implementation challenges, and future directions. Results: Whole-genome sequencing is increasingly being adopted as a first-line genomic test for suspected rare genetic disorders, particularly when the phenotype is heterogeneous or does not point to a specific diagnosis. Conventional cytogenetic and targeted molecular techniques remain important complementary approaches for selected phenotypes, variant classes, and orthogonal confirmation. Gene panels are effective for well-defined phenotypes, whereas whole-exome sequencing remains a high-yield approach for genetically heterogeneous disorders, particularly when whole-genome sequencing is not available or is not clinically indicated. Long-read whole-genome sequencing expands diagnostic capacity by detecting structural variants, repeat expansions, complex rearrangements, and non-coding pathogenic variants that frequently escape short-read technologies. Emerging multi-omics approaches further improve variant interpretation and help resolve previously unsolved cases. Beyond diagnosis, molecular findings guide personalized clinical management, genetic counselling, reproductive planning, and access to targeted therapies and genotype-driven clinical trials. However, broad implementation is constrained by challenges in variant interpretation, ethical and legal considerations, data governance, workforce capacity, cost, and inequitable access to genomic services. Artificial intelligence, international data-sharing initiatives, and coordinated healthcare networks are helping overcome these barriers and improve diagnostic equity. Conclusions: Whole-genome sequencing is increasingly emerging as a first-line genomic strategy for pediatric rare diseases, while complementary technologies, expert phenotyping, and iterative data interpretation remain essential for comprehensive and accurate diagnosis and equitable access to genomic medicine. Full article

Introduction: Dating violence is an early form of intimate partner violence characterized by controlling behaviors, psychological and sexual abuse, and digital coercion. Despite its high prevalence among adolescents and young adults, it remains underrecognized and insufficiently addressed in healthcare settings. Nurses are strategically positioned to identify and respond to dating violence due to their close contact with patients and holistic approach to care. However, important gaps remain between existing protocols and their implementation in clinical practice. Methods: An exploratory qualitative study was conducted within a constructivist paradigm as part of a multicenter research project on dating violence among health sciences university students. Twelve experts, including nurses, healthcare professionals, and representatives of survivor-led organizations, participated in focus groups and reflective journals between September and December 2025 at the University of Barcelona. Data were analyzed using an interpretative phenomenological approach, with inductive thematic analysis guiding coding and theme development. Methodological rigor was ensured following the Consolidated Criteria for Reporting Qualitative Research (COREQ). Ethical approval was obtained. Results: Four themes emerged: (1) barriers to identification, including the normalization of controlling behaviors, myths of romantic love, and limited recognition of psychological and digital violence; (2) challenges in nursing responses, such as insufficient training, clinical uncertainty, organizational barriers, and lack of screening tools; (3) identification strategies, including systematic screening, relationship-centered interviewing, emotional education, and recognition of subtle warning signs; and (4) actions following identification, focused on validation, confidentiality, risk assessment, interdisciplinary referral, and ongoing support. Conclusions: Findings highlight the need to update protocols to explicitly address digital violence and strengthen nursing-led interventions. Participants emphasized the need for enhanced professional training, context-appropriate screening tools, and intersectoral collaboration as potential strategies to improve the identification and response to dating violence. Full article

Artificial intelligence (AI) is developing rapidly and is gradually transforming traditional management control practices into intelligent management control, characterized by automation, analysis and decision-making. This research aims to explore this transition from traditional management control, focused on retrospective monitoring, to intelligent management control, by highlighting two further paradigms: predictive and ethical management control. The methodology is based on semi-structured interviews with Moroccan management controllers working in various sectors. These interviews provided in-depth insights into their experiences of using AI tools, the perceived benefits and the limitations encountered in their use. The results highlight that artificial intelligence makes several significant contributions to the work of management controllers, improving the efficiency of analyses, the quality of information and the speed of decision-making processes, thereby allowing them to focus on activities with higher added value. However, the study also highlights the need to develop new capabilities in prompt engineering, strategic communication and the management of human–machine relationships, in order to adapt to these transformations. Nevertheless, this evolution is facilitating a gradual repositioning of the management controller towards a more strategic role. Furthermore, it documents the specific characteristics of technology adoption in the African context, which is characterized by the phenomenon of shadow AI. Full article