Search Results (144)

Adverse drug events (ADEs) significantly impact patient safety and health outcomes. Manual ADE detection from clinical narratives is time-consuming, labor-intensive, and costly. Recent advancements in large language models (LLMs), including transformer-based architectures such as Bidirectional Encoder Representations from Transformers (BERT) and Generative Pretrained Transformer (GPT) series, offer promising methods for automating ADE extraction from clinical data. These models have been applied to various aspects of pharmacovigilance and clinical decision support, demonstrating potential in extracting ADE-related information from real-world clinical data. Additionally, chatbot-assisted systems have been explored as tools in clinical management, aiding in medication adherence, patient engagement, and symptom monitoring. This narrative review synthesizes the current state of LLMs in ADE detection from a clinical perspective, organizing studies into categories such as human-facing decision support tools, immune-related ADE detection, cancer-related and non-cancer-related ADE surveillance, and personalized decision support systems. In total, 39 articles were included in this review. Across domains, LLM-driven methods have demonstrated promising performances, often outperforming traditional approaches. However, critical limitations persist, such as domain-specific variability in model performance, interpretability challenges, data quality and privacy concerns, and infrastructure requirements. By addressing these challenges, LLM-based ADE detection could enhance pharmacovigilance practices, improve patient safety outcomes, and optimize clinical workflows. Full article

The growing influence of technology in the healthcare industry has led to the creation of innovative applications that improve convenience, accessibility, and diagnostic accuracy. However, health applications face significant challenges concerning user privacy and data security, as they handle extremely sensitive personal and medical information. Privacy-Enhancing Technologies (PETs), such as Privacy-Attribute-based Credentials, Differential Privacy, and Federated Learning, have emerged as crucial tools to tackle these challenges. Despite their potential, PETs are not widely utilized due to technical and implementation obstacles. This research introduces a comprehensive framework for protecting health applications from privacy and security threats, with a specific emphasis on gamified mental health apps designed to manage Attention Deficit Hyperactivity Disorder (ADHD) in children. Acknowledging the heightened sensitivity of mental health data, especially in applications for children, our framework prioritizes user-centered design and strong privacy measures. We suggest an identity management system based on blockchain technology to ensure secure and transparent credential management and incorporate Federated Learning to enable privacy-preserving AI-driven predictions. These advancements ensure compliance with data protection regulations, like GDPR, while meeting the needs of various stakeholders, including children, parents, educators, and healthcare professionals. Full article

Patient-generated health data (PGHD) enhance traditional healthcare by enabling continuous monitoring and supporting personalized care, yet concerns over privacy, security, and integration into existing systems hinder broader adoption. This study examined the perceptions, awareness, and concerns of patients and caregivers regarding PGHD and assessed their willingness to share such data for clinical, research, and commercial purposes. A cross-sectional survey was conducted from 6 to 12 November 2023, involving 400 individuals with experience using PGHD. Participants completed structured questionnaires addressing health information management, PGHD usage, and attitudes toward its application. PGHD was most commonly used by patients with chronic conditions and guardians of minors, with tethered personal health record apps frequently utilized. Respondents identified improved self-management and better access to information as key benefits. However, significant concerns about data privacy and security emerged, especially regarding non-clinical use. Younger adults, particularly those in their 20s, showed lower willingness to engage with PGHD due to heightened privacy concerns. These findings suggest that, while support for clinical use of PGHD is strong, barriers related to trust and consent remain. Addressing privacy concerns and simplifying consent processes will be essential to promote equitable and responsible PGHD utilization across diverse patient populations. Full article

Large language models (LLMs) are transforming the capabilities of medical chatbots by enabling more context-aware, human-like interactions. This review presents a comprehensive analysis of their applications, technical foundations, benefits, challenges, and future directions in healthcare. LLMs are increasingly used in patient-facing roles, such as symptom checking, health information delivery, and mental health support, as well as in clinician-facing applications, including documentation, decision support, and education. However, as a study from 2024 warns, there is a need to manage “extreme AI risks amid rapid progress”. We examine transformer-based architectures, fine-tuning strategies, and evaluation benchmarks specific to medical domains to identify their potential to transfer and mitigate AI risks when using LLMs in medical chatbots. While LLMs offer advantages in scalability, personalization, and 24/7 accessibility, their deployment in healthcare also raises critical concerns. These include hallucinations (the generation of factually incorrect or misleading content by an AI model), algorithmic biases, privacy risks, and a lack of regulatory clarity. Ethical and legal challenges, such as accountability, explainability, and liability, remain unresolved. Importantly, this review integrates broader insights on AI safety, drawing attention to the systemic risks associated with rapid LLM deployment. As highlighted in recent policy research, including work on managing extreme AI risks, there is an urgent need for governance frameworks that extend beyond technical reliability to include societal oversight and long-term alignment. We advocate for responsible innovation and sustained collaboration among clinicians, developers, ethicists, and regulators to ensure that LLM-powered medical chatbots are deployed safely, equitably, and transparently within healthcare systems. Full article

Human Activity Recognition (HAR) is crucial for understanding human behaviour through sensor data, with applications in healthcare, smart environments, and surveillance. While traditional HAR often relies on ambient sensors, wearable devices or vision-based systems, these approaches can face limitations in dynamic settings and raise privacy concerns. Device-free HAR systems, utilising Wi-Fi Channel State Information (CSI) to human movements, have emerged as a promising privacy-preserving alternative for next-generation health activity monitoring and smart environments, particularly for multi-user scenarios. However, current research faces challenges such as the need for substantial annotated training data, class imbalance, and poor generalisability in complex, multi-user environments where labelled data is often scarce. This paper addresses these gaps by proposing a hybrid deep learning approach which integrates signal preprocessing, targeted data augmentation, and a customised integration of CNN and Transformer models, designed to address the challenges of multi-user recognition and data scarcity. A random transformation technique to augment real CSI data, followed by hybrid feature extraction involving statistical, spectral, and entropy-based measures to derive suitable representations from temporal sensory input, is employed. Experimental results show that the proposed model outperforms several baselines in single-user and multi-user contexts. Our findings demonstrate that combining real and augmented data significantly improves model generalisation in scenarios with limited labelled data. Full article

Background: The rapid integration of artificial intelligence (AI) technologies into healthcare systems presents new opportunities and challenges, particularly regarding legal and ethical implications. In Saudi Arabia, the lack of legal awareness could hinder safe implementation of AI tools. Methods: A sequential explanatory mixed-methods design was employed. In Phase One, a structured electronic survey was administered to 357 clinicians across public and private healthcare institutions in Saudi Arabia, assessing legal awareness, liability concerns, data privacy, and trust in AI. In Phase Two, a qualitative expert panel involving health law specialists, digital health advisors, and clinicians was conducted to interpret survey findings and identify key regulatory needs. Results: Only 7% of clinicians reported high familiarity with AI legal implications, and 89% had no formal legal training. Confidence in AI compliance with data laws was low (mean score: 1.40/3). Statistically significant associations were found between professional role and legal familiarity (χ² = 18.6, p < 0.01), and between legal training and confidence in AI compliance (t ≈ 6.1, p < 0.001). Qualitative findings highlighted six core legal barriers including lack of training, unclear liability, and gaps in regulatory alignment with national laws like the Personal Data Protection Law (PDPL). Conclusions: The study highlights a major gap in legal readiness among Saudi clinicians, which affects patient safety, liability, and trust in AI. Although clinicians are open to using AI, unclear regulations pose barriers to safe adoption. Experts call for national legal standards, mandatory training, and informed consent protocols. A clear legal framework and clinician education are crucial for the ethical and effective use of AI in healthcare. Full article

Adolescents living with HIV (ALHIV) face significant challenges in self-managing their chronic condition. Digital health technology (DHT) has become increasingly common and understanding ALHIVs’ preferences is essential for developing interventions tailored to this unique population. This study aimed to explore the preferences of ALHIV regarding the use of DHT for self-management. A qualitative research approach with an exploratory and descriptive design was used. Participants were recruited using a purposive sampling method. Data were gathered through six nominal focus groups with 29 participants at two Community Health Centers in the Western Cape Province, South Africa. The participants were ALHIV aged 15–24 years. Discussions focused on current technology usage and the ranking of desired DHT features. The transcripts were analyzed using thematic analysis. Three main themes emerged: (1) everyday usage of digital technology where participants frequently used digital devices for communication, social media, and finding information; (2) the role of digital technology in self-management; a strong interest in digital technology that provides medication reminders, health education, and peer support; and (3) factors influencing digital technology, including the cost of data, limited connectivity, and issues of privacy related to participants’ HIV status. The ALHIV showed a strong willingness to use digital platforms for health information, reminders, and peer support, although concerns about connectivity, data cost, and privacy remain. These findings underscore the need for flexible, user-centered approaches when designing DHT interventions for self-management in South Africa. Full article

Background and Objectives: AI is transforming healthcare, with family doctors at the forefront. As primary care providers, they play a key role in integrating AI into patient care. Despite AI’s potential, concerns about trust, data privacy, and physician autonomy persist. Little research exists on family doctors’ perspectives. This study investigates the views of Lithuanian family physicians on AI’s ethical challenges and benefits, aiming to support responsible implementation. Materials and Methods: A review of the literature was conducted (2015–2025) using Google Scholar, PubMed, and Scopus. This qualitative study explored family physicians’ perceptions of AI in Lithuania, focusing on ethics, AI’s role, experience, training, and concerns about replacement. Informed consent and ethical guidelines were followed. Results: AI has strong potential in family medicine, automating administrative tasks, improving diagnostic accuracy, and supporting patient autonomy. AI tools, like clinical documentation systems and smart devices save time, allowing physicians to focus on patient care. They also improve diagnostic precision, enabling earlier detection of conditions such as cancer and coronary artery disease. Physicians express concerns about AI’s reliability, biases, and data privacy. While AI boosts efficiency, many emphasize the importance of human oversight in decision-making, especially in complex cases. Privacy concerns around health data and the need for stricter regulations are crucial. Lithuanian family physicians generally accept AI as a helpful tool for routine tasks but remain cautious regarding its trustworthiness. Job displacement concerns were not prevalent, with AI seen as a tool to augment rather than replace their role. Successful AI integration requires training, transparency, and ethical guidelines to build trust and ensure patient safety. Conclusions: AI enhances efficiency in family medicine but requires structured training and ethical safeguards to address concerns about data privacy, accountability, and bias. AI is viewed as supportive, not as a replacement. Full article

Background: The increasing prevalence of dementia and mild cognitive impairment (MCI) among the elderly population is a global health issue. Information and Communication Technology (ICT)-based interventions hold promises for maintaining cognition, but their viability is affected by several challenges. Objectives: This study aimed to significantly assess the effectiveness of ICT-based cognitive and memory aid technology for individuals with MCI or dementia, identify adoption drivers, and develop an implementation model to inform practice. Methods: A PRISMA-based systematic literature review, with the protocol registered in PROSPERO (CRD420251051515), was conducted using seven electronic databases published between January 2015 and January 2025 following the PECOS framework. Random effects models were used for meta-analysis, and risk of bias was assessed using the Joanna Briggs Institute (JBI) Critical Appraisal Checklists. Results: A total of ten forms of ICT interventions that had proved effective to support older adults with MCI and dementia. Barriers to adoption included digital literacy differences, usability issues, privacy concerns, and the lack of caregiver support. Facilitators were individualized design, caregiver involvement, and culturally appropriate implementation. ICT-based interventions showed moderate improvements in cognitive outcomes (pooled Cohen’s d = 0.49, 95% CI: 0.14–1.03). A sensitivity analysis excluding high-risk studies yielded a comparable effect size (Cohen’s d = 0.50), indicating robust findings. However, trim-and-fill analysis suggested a slightly reduced corrected effect (Cohen’s d = 0.39, 95% CI: 0.28–0.49), reflecting potential small-study bias. Heterogeneity was moderate (I² = 46%) and increased to 55% after excluding high-risk studies. Subgroup analysis showed that tablet-based interventions tended to produce higher effect sizes. Conclusions: ICT-based interventions considerably enhance cognition status, autonomy, and social interaction in older adults with MCI and dementia. To ensure long-term scalability, future initiatives must prioritize user-centered design, caregiver education, equitable access to technology, accessible infrastructure and supportive policy frameworks. Full article

Personal health records (PHRs) are digital repositories that allow individuals to access, manage, and share their health information. By enabling patients to track their health over time and communicate effectively with healthcare providers, personal health records support more personalized care and improve the quality of healthcare. Their integration with emerging technologies, such as the Internet of Things (IoT), artificial intelligence (AI), and blockchain, enhances the utility and security of health data management, facilitating continuous health monitoring, automated decision support, and secure, decentralized data exchange. Despite their potential, PHR systems face significant challenges, including privacy concerns, security issues, and digital accessibility problems. This paper discusses the fundamental concepts, requirements, system architectures, and data sources that underpin modern PHR implementations, highlighting how they enable continuous health monitoring through the integration of wearable sensors; mobile health applications; and IoT-enabled medical devices that collect, process, and transmit data to support proactive care and personalized treatments. The benefits and limitations of PHR systems are also discussed in detail, with a focus on interoperability, adoption drivers, and the role of advanced technologies in supporting the development of secure and scalable health information systems for a smarter world. Full article

Purpose: Assess mental health professionals’ attitudes regarding the timing and characteristics of therapeutic interventions for children whose parents have incurable cancer, and whether professionals would use artificial intelligence (AI) in these interventions. Methods: Professionals were surveyed about their therapeutic approaches to caring for children when parents have incurable cancer under different scenarios. Data from N = 294 (69% male, 72% white, 26% Latine, 56% rural or underserved communities) physicians, psychologists, social workers, hospital chaplains, community health workers, and others were analyzed. Attitudes surrounding the timing and characteristics of interventions across the parent’s cancer journey were compared, including how professionals believed interventions should attend to dimensions of the child or family, and if, how, and when AI technology could be introduced. Results: Across 10 dimensions of childhood, (1) the child’s premorbid exposure to traumatic events, (2) a surviving parent’s presence, and (3) the child’s age were important factors to consider when making mental health care decisions in this context. The professionals reported being more likely to introduce therapeutic resources as early as possible in the parent’s illness (i.e., upon diagnosis). Regarding the use of AI, 87% foresaw its role in supporting children’s mental health. While 93.2% agreed that a grieving child could be helped by interacting with an AI-generated likeness of the deceased parent, when AI’s use was contextualized in providing support for a child who lost a parent to cancer, only 49% believed AI was appropriate. The participants were conflicted over when AI could be first introduced, either upon a parent’s illness diagnosis (19.4%), during a parent’s treatment (19.0%), or as part of a parent’s hospice care (12.6%). None believed it to be appropriate following the loss of the parent to cancer. Conclusions: AI is increasingly present in children’s daily lives and quickly infiltrating health care with widely accessible mental health chatbots. Concerns about privacy, the accuracy of information, and the anthropomorphism of AI tools by children give professionals pause before introducing such technology. Proceeding with great caution is urged until more is known about the impact of AI on children’s mental health, grief, and psychological well-being in the context of parental cancer. Full article

“Sharenting” appears to have become a common practice among families, who tend to normalise the posting of children’s content on social media, which can raise concerns about the privacy, safety, and mental health of exposed children. This study examines the perceptions and practices of sharenting among families in Asunción (Paraguay). A survey of 73 parents analysed posting habits, knowledge of risks, and possible influencing factors on parental digital behaviour. Data analysis techniques such as descriptive statistics and correlation analysis were used to examine the associations between the key variables. The results reveal that 72.60% of respondents publish content about their children on social networks, while 95.89% recognise that they are concerned about the risks associated with this practice. In addition, 58.90% of the participants indicated that they were unaware of the term sharenting. The analysis suggests that there is no significant association between knowing one’s social media contacts and the decision to post information about one’s children, indicating that perceived privacy may not directly influence the practice of sharenting. This highlights the need to educate families and promote awareness of the risks of children’s exposure to digital platforms. Full article

Data generated by Internet of Things devices enable the design of new business models and services, improving user experience and satisfaction. This data also serve as an essential information source for many fields, including disaster management, bio-surveillance, smart cities, and smart health. However, personal data are also collected in this context, introducing new challenges concerning data privacy protection, such as profiling, localization and tracking, linkage, and identification. This dilemma is further complicated by the “privacy paradox”, where users compromise privacy for service convenience. Hence, this paper reviews the literature on data privacy in the IoT, particularly emphasizing Personal Data Store (PDS)-based approaches as a promising class of user-centric solutions. PDS represents a user-centric approach to decentralizing data management, enhancing privacy by granting individuals control over their data. Addressing privacy solutions involves a triad of user privacy awareness, technology support, and ways to regulate data processing. Our discussion aims to advance the understanding of IoT privacy issues while emphasizing the potential of PDS to balance privacy protection and service delivery. Full article

As populations age globally, governments face mounting challenges in reconfiguring healthcare and housing systems to support aging-in-place. This study offers a comparative analysis of South Korea and the United Kingdom, examining how each country integrates digital technologies, such as Artificial Intelligence (AI), telecare, and smart housing systems, into their aging strategies. South Korea employs a centralized, technology-driven approach that prioritizes the national rollout of AI-enabled smart homes and digital health infrastructure. In contrast, the UK advances a decentralized, community-based model emphasizing social housing, localized care delivery, and telecare integration. Despite these differing trajectories, both nations face shared limitations, including high implementation costs, digital literacy barriers, and concerns about data privacy. Critically, the study finds that the success of aging-in-place efforts is shaped not only by technological capacity but also by governance dynamics, political continuity, and institutional coordination. In response, the paper proposes policy recommendations alongside an ethical framework grounded in transparency, autonomy, informed consent, and equity. Sustainable aging-in-place strategies require not only innovative technologies, but also inclusive governance and ethically robust design to ensure accessibility, trust, and long-term impact. Full article

Medical Internet of Things (IoT) systems are crucial in monitoring the health status of patients. Recently, telemedicine services that manage patients remotely by receiving real-time health information from IoT devices attached to or carried by them have experienced significant growth. A primary concern in medical IoT services is ensuring the security of transmitted information and protecting patient privacy. To address these challenges, various authentication schemes have been proposed. We analyze the authentication scheme by Wang et al. and identified several limitations. Specifically, an attacker can exploit information stored in an IoT device to generate an illegitimate session key. Additionally, despite using a cloud center, the scheme lacks efficiency. To overcome these limitations, we propose an authentication and key distribution scheme that incorporates a physically unclonable function (PUF) and public-key computation. To enhance efficiency, computationally intensive public-key operations are performed exclusively in the cloud center. Furthermore, our scheme addresses privacy concerns by employing a temporary ID for IoT devices used to identify patients. We validate the security of our approach using the formal security analysis tool ProVerif. Full article