Rachel C Stockley, Yasemin Hirst, Chantelle Hayes, Kimberley E Watkins, Peter C Goodwin
{"title":"How equitable is digital rehabilitation for people after stroke? A systematic review using an equity approach.","authors":"Rachel C Stockley, Yasemin Hirst, Chantelle Hayes, Kimberley E Watkins, Peter C Goodwin","doi":"10.3389/fdgth.2025.1544754","DOIUrl":"10.3389/fdgth.2025.1544754","url":null,"abstract":"<p><strong>Introduction: </strong>Stroke is the largest global cause of adult neuro-disability. Health inequities increase the risk of stroke and are likely to influence overall recovery. Rehabilitation after stroke seeks to restore function and independence and may utilise digital technologies to augment usual care. This study systematically investigates the reporting of equity factors in digital stroke rehabilitation research.</p><p><strong>Methods: </strong>This systematic review examined equity factors contained in the PROGRESS-Plus framework in a random sample of clinical trials of technologies used as part of stroke rehabilitation published in 2011-2021. Four reviewers double-screened titles and abstracts of 14,724 papers. A random selection was carried out across all potentially eligible papers (<i>n</i> = 821) and 135 papers were reviewed for data extraction. Each study was coded with 36-point PROGRESS-plus criteria for inclusion, exclusion, and baseline characteristics. ANOVA and multivariable linear regression were used to assess the variation in PROGRESS-Plus reporting by year of publication, location, type of technology used, intervention target, number of comparison groups and sample size.</p><p><strong>Results: </strong>87 studies were included with a mean PROGRESS-Plus score of 7.05 (<i>SD</i> <i>=</i> 2.06), minimum score of 0 and maximum score of 14. Despite their importance to health outcomes, education, social capital and socioeconomic status were reported by less than 5% of studies. The most commonly reported equity factors were age, disability and gender. There were no significant differences in reporting by technology used, target of the intervention (upper or lower limb), sample size, location, number of comparison groups and sample size. Variation in equity reporting was not explained through multiple linear regression factors. There was a small positive correlation between the year of publication and the PROGRESS-Plus score (<i>r</i> = .26, <i>n</i> = 87, <i>p</i> < 0.05).</p><p><strong>Discussion: </strong>Few studies of digital rehabilitation interventions considered several key equity factors, including those recognised to precipitate digital exclusion and influence health outcomes. An encouraging finding was that more recent work was slightly more likely to report equity factors, but future research should ensure complete reporting of equity factors to ensure their findings are applicable to clinical populations.</p><p><strong>Systematic review registration: </strong>https://www.crd.york.ac.uk/PROSPERO/view/CRD42024504300, PROSPERO/identifier, CRD42024504300.</p>","PeriodicalId":73078,"journal":{"name":"Frontiers in digital health","volume":"7 ","pages":"1544754"},"PeriodicalIF":3.2,"publicationDate":"2025-06-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12234467/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144593038","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"A novel personal identification system using doorknob lead electrocardiograms for unconscious authentication in unlocking doors.","authors":"Keisuke Kawamura, Masaki Kyoso","doi":"10.3389/fdgth.2025.1585431","DOIUrl":"10.3389/fdgth.2025.1585431","url":null,"abstract":"<p><strong>Introduction: </strong>In highly information-oriented society, personal authentication technology is essential. Biometric authentication is becoming popular as a method of personal authentication from the viewpoint of usability. In this research, in order to realize unconscious personal authentication during daily activities, we proposed a novel biometric authentication system using a doorknob-type electrocardiogram (ECG) measuring device. In our previous study, it was shown that ECG obtained with a contact-type electrode on doorknob and a capacitive-type electrode on the floor could be used for personal identification. However, identification performance is easily affected by noise from body movements and other factors, due to loose contact between electrodes and the body.</p><p><strong>Method: </strong>In this paper, we proposed to add two preprocessing techniques to the system. Synchronized averaging process was applied to the measured ECG waveforms. Then, data augmentation was applied to the machine learning training data.</p><p><strong>Results: </strong>It was found that synchronized averaging with 5 consecutive wave segment improved accuracy by 10%. It was also found that training data augmentation improved the performance even under limited amount of ECG data.</p><p><strong>Discussion: </strong>The results demonstrate that remarkable performance improvement can be achieved even with short term door-knob ECG by using synchronized averaging and data augmentation.</p>","PeriodicalId":73078,"journal":{"name":"Frontiers in digital health","volume":"7 ","pages":"1585431"},"PeriodicalIF":3.2,"publicationDate":"2025-06-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12226577/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144577112","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Rosie Stenhouse, Wassie Gebbie Beshir, Demessie Girma, Gosaye Fida, Clara Calia, Godana Guto, Maria Klara Wolters
{"title":"Developing a COVID-19-focused mHealth system in a low-resource setting during the COVID-19 pandemic: challenges and opportunities.","authors":"Rosie Stenhouse, Wassie Gebbie Beshir, Demessie Girma, Gosaye Fida, Clara Calia, Godana Guto, Maria Klara Wolters","doi":"10.3389/fdgth.2025.1543828","DOIUrl":"10.3389/fdgth.2025.1543828","url":null,"abstract":"<p><strong>Introduction: </strong>Approximately three-quarters of Ethiopia's population lives in rural areas, and access to healthcare is difficult with poor transport infrastructure and long travel times. Telemedicine has the potential to support healthcare access and minimise COVID-19 transmission through a reduced need to travel.</p><p><strong>Objectives: </strong>This Brief Research Report describes the analysis of qualitative data relating to the development of a mobile health (mHealth) system during the COVID-19 pandemic to support COVID-19 symptom management in the community in Oromia, Ethiopia.</p><p><strong>Methods: </strong>Data were collected from (1) meeting notes and WhatsApp group discussions, (2) a focus group with medical staff, and (3) an interview with a senior hospital leader. A framework method was used for the analysis.</p><p><strong>Results: </strong>Three themes were identified: (1) patient-physician relationship, (2) new ways of using everyday technology, and (3) infrastructure and digital access.</p><p><strong>Discussion: </strong>We discuss the challenges of developing an mHealth system during a pandemic alongside infrastructural challenges and the preparedness of medical staff and communities for the use of mHealth.</p><p><strong>Conclusions: </strong>There is a need for investment in information technology infrastructure and in access to digital networks, alongside a need to improve the digital and health literacy of populations for the successful implementation of a patient-facing mHealth system. Thus, whilst the policy aspirations are admirable, the potential for technological innovation is great, and the clinicians can see the benefit of using technologies to provide care to those who cannot reach clinics, there is a gap between what is possible given the current reality of infrastructure and patient preparedness and the requirements for a successful telemedicine intervention.</p>","PeriodicalId":73078,"journal":{"name":"Frontiers in digital health","volume":"7 ","pages":"1543828"},"PeriodicalIF":3.2,"publicationDate":"2025-06-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12222052/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144562235","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Imanol Pinto, Álvaro Olazarán, David Jurío, Borja De la Osa, Miguel Sainz, Aritz Oscoz, Jerónimo Ballaz, Javier Gorricho, Mikel Galar, José Andonegui
{"title":"Improving diabetic retinopathy screening using artificial intelligence: design, evaluation and before-and-after study of a custom development.","authors":"Imanol Pinto, Álvaro Olazarán, David Jurío, Borja De la Osa, Miguel Sainz, Aritz Oscoz, Jerónimo Ballaz, Javier Gorricho, Mikel Galar, José Andonegui","doi":"10.3389/fdgth.2025.1547045","DOIUrl":"10.3389/fdgth.2025.1547045","url":null,"abstract":"<p><strong>Background: </strong>The worst outcomes of diabetic retinopathy (DR) can be prevented by implementing DR screening programs assisted by AI. At the University Hospital of Navarre (HUN), Spain, general practitioners (GPs) grade fundus images in an ongoing DR screening program, referring to a second screening level (ophthalmologist) target patients.</p><p><strong>Methods: </strong>After collecting their requirements, HUN decided to develop a custom AI tool, called NaIA-RD, to assist their GPs in DR screening. This paper introduces NaIA-RD, details its implementation, and highlights its unique combination of DR and retinal image quality grading in a single system. Its impact is measured in an unprecedented before-and-after study that compares 19,828 patients screened before NaIA-RD's implementation and 22,962 patients screened after.</p><p><strong>Results: </strong>NaIA-RD influenced the screening criteria of 3/4 GPs, increasing their sensitivity. Agreement between NaIA-RD and the GPs was high for non-referral proposals (94.6% or more), but lower and variable (from 23.4% to 86.6%) for referral proposals. An ophthalmologist discarded a NaIA-RD error in most of contradicted referral proposals by labeling the 93% of a sample of them as referable. In an autonomous setup, NaIA-RD would have reduced the study visualization workload by 4.27 times without missing a single case of sight-threatening DR referred by a GP.</p><p><strong>Conclusion: </strong>DR screening was more effective when supported by NaIA-RD, which could be safely used to autonomously perform the first level of screening. This shows how AI devices, when seamlessly integrated into clinical workflows, can help improve clinical pathways in the long term.</p>","PeriodicalId":73078,"journal":{"name":"Frontiers in digital health","volume":"7 ","pages":"1547045"},"PeriodicalIF":3.2,"publicationDate":"2025-06-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12222278/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144562237","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Sarah van Drumpt, Kartik Chawla, Tom Barbereau, Dayana Spagnuelo, Linda van de Burgwal
{"title":"Secondary use under the European Health Data Space: setting the scene and towards a research agenda on privacy-enhancing technologies.","authors":"Sarah van Drumpt, Kartik Chawla, Tom Barbereau, Dayana Spagnuelo, Linda van de Burgwal","doi":"10.3389/fdgth.2025.1602101","DOIUrl":"10.3389/fdgth.2025.1602101","url":null,"abstract":"<p><p>The Regulation for European Health Data Space (EHDS) aims to address the fragmented health data landscape across Europe by promoting ethical and responsible reuse of data, seeking to balance the opportunities for data reuse with the risks it entails. However, the techno-legal aspects of navigating this balance remain poorly understood. This study adopts a qualitative and inductive approach, using semi-structured interviews to explore the risks, challenges, and gaps in the implementation of privacy-enhancing technologies (PETs) within EHDS, particularly in the context of its governance structure and data permits for secondary data use. The findings identify five distinct categories of concerns, based on fourteen risks, and highlight seven governance and technological solutions, illustrating how these solutions address multiple, often correlated risks. The interdependence between concerns and solutions emphasises the need for a strategic and integrated approach to both governance and technology. This mapping between the risks and solutions also highlights the central role of certain solutions, such as public engagement and awareness, in addressing multiple risks. Furthermore, it introduces a new dimension to the concerns by focusing on the structural imbalances in access to the health data economy. We conclude by proposing a research agenda to advance the integration of PETs into the EHDS framework, ensuring that data permits can effectively facilitate secure, ethical, and innovative health data use.</p>","PeriodicalId":73078,"journal":{"name":"Frontiers in digital health","volume":"7 ","pages":"1602101"},"PeriodicalIF":3.2,"publicationDate":"2025-06-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12222193/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144562239","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Enhancing heart disease prediction with stacked ensemble and MCDM-based ranking: an optimized RST-ML approach.","authors":"T Ashika, G Hannah Grace","doi":"10.3389/fdgth.2025.1609308","DOIUrl":"10.3389/fdgth.2025.1609308","url":null,"abstract":"<p><strong>Introduction: </strong>Cardiovascular disease (CVD) is a leading global cause of death, necessitating the development of accurate diagnostic models. This study presents an Optimized Rough Set Theory-Machine Learning (RST-ML) framework that integrates Multi-Criteria Decision-Making (MCDM) for effective heart disease (HD) prediction. By utilizing RST for feature selection, the framework minimizes dimensionality while retaining essential information.</p><p><strong>Methods: </strong>The framework employs RST to select relevant features, followed by the integration of nine ML classifiers into five stacked ensemble models through correlation analysis to enhance predictive accuracy and reduce overfitting. The Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) ranks the models, with weights assigned using the Mean Rank Error Correction (MEREC) method. Hyperparameter tuning for the top model, Stack-4, was conducted using GridSearchCV, identifying XGBoost (XG) as the most effective classifier. To assess scalability and generalization, the framework was evaluated using additional datasets, including chronic kidney disease (CKD), obesity levels, and breast cancer. Explainable AI (XAI) techniques were also applied to clarify feature importance and decision-making processes.</p><p><strong>Results: </strong>Stack-4 emerged as the highest-performing model, with XGBoost achieving the best predictive accuracy. The application of XAI techniques provided insights into the model's decision-making, highlighting key features influencing predictions.</p><p><strong>Discussion: </strong>The findings demonstrate the effectiveness of the RST-ML framework in improving HD prediction accuracy. The successful application to diverse datasets indicates strong scalability and generalization potential, making the framework a robust and scalable solution for timely diagnosis across various health conditions.</p>","PeriodicalId":73078,"journal":{"name":"Frontiers in digital health","volume":"7 ","pages":"1609308"},"PeriodicalIF":3.2,"publicationDate":"2025-06-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12222165/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144562236","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Beyond the interface: benchmarking pediatric mobile health applications for monitoring child growth using the Mobile App Rating Scale.","authors":"Anggi Septia Irawan, Arie Dwi Alristina, Rizky Dzariyani Laili, Nuke Amalia, Arief Purnama Muharram, Adriana Viola Miranda, Bence Döbrössy, Edmond Girasek","doi":"10.3389/fdgth.2025.1621293","DOIUrl":"10.3389/fdgth.2025.1621293","url":null,"abstract":"<p><strong>Introduction: </strong>As mHealth applications become increasingly adopted in Indonesia, it is crucial to assess their quality and usability for parents and healthcare professionals.</p><p><strong>Aim: </strong>This study evaluated the quality of pediatric-related mobile health (mHealth) applications available in Indonesia, focusing on their ability to support child growth monitoring and provide educational resources for parents and caregivers.</p><p><strong>Methodology: </strong>This is a cross-sectional study. From December 1, 2024, and January 31, 2025 we conducted systematic search for pediatric mHealth applications in Indonesian Google Play Store and Apple App Store using predetermined keywords. Inclusion criteria required the applications to be available in Bahasa Indonesia, focus on child health, and include growth tracking or stunting prevention features. We excluded applications that were not functioning during the testing period. Quality assessment was conducted by five healthcare professionals using the Mobile App Rating Scale (MARS). MARS assessed applications from multiple domains, including engagement, functionality, aesthetics, and information quality. Inter-rater reliability was ensured using the Intraclass Correlation Coefficient (ICC). The results were analyzed using descriptive statistics, Pearson's correlation, and T-tests. A <i>p</i>-value of <0.05 is considered to be statistically significant.</p><p><strong>Findings: </strong>Nine applications were included in this study. Seven of the applications (77.78%) focused on tracking child growth and development and providing educational content. Less than half of the apps had built-in community features that enabled social support (<i>n</i> = 4, 44.44%) and features for feedback mechanisms & personalized guidance (<i>n</i> = 3, 33.33%) respectively. The majority were developed by commercial companies (<i>n</i> = 7, 77.78%). Quality assessment found significant variability across the apps, with high functionality and aesthetics scores but more variability in the domains of app engagement, quality of information, and subjective quality or perceived value.</p><p><strong>Conclusion: </strong>This research underscored the need for the development of higher-quality, evidence-based mHealth apps for pediatric care in Indonesia, particularly in improving user engagement, feedback mechanisms and accessibility.</p>","PeriodicalId":73078,"journal":{"name":"Frontiers in digital health","volume":"7 ","pages":"1621293"},"PeriodicalIF":3.2,"publicationDate":"2025-06-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12213656/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144556075","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Research trends in the application of artificial intelligence in nursing of chronic disease: a bibliometric and network visualization study.","authors":"Chao Du, Jing Zhou, Yuexin Yu","doi":"10.3389/fdgth.2025.1608266","DOIUrl":"10.3389/fdgth.2025.1608266","url":null,"abstract":"<p><strong>Purpose: </strong>The incidence of chronic diseases is increasing annually and exhibits a trend of multimorbidity, posing significant challenges to global healthcare and nursing. The rapid rise of artificial intelligence has provided broad application prospects in the field of chronic disease care. However, with the increasing number of related studies, there is a lack of systematic review and prediction of future trends in this area. Bibliometric methods provide possibility for addressing this gap. This study aimed to investigate the current status, hot topics, and future prospects of artificial intelligence in the field of chronic disease care.</p><p><strong>Methods: </strong>Literature related to artificial intelligence and chronic disease care was retrieved from the Web of Science Core Collection database, published between 2001 and 31 December 2023. Bibliometric analysis and visualization was conducted using CiteSpace 5.7.R5 and VOSviewer 1.6.19 to analyze countries/regions, institutions, journals, references, and keywords.</p><p><strong>Results: </strong>A total of 2438 articles were retrieved, indicating an explosive growth in publications over the past five years. The United States emerged as the earliest adopter of research in this domain (since 2002) and contributed the most publications (490 articles), with IEEE ACCESS being the most cited journal. Hot application areas of artificial intelligence in chronic disease care included \"diabetic retinopathy\", \"heart disease prediction\", \"breast cancer\", and \"skin cancer\". Major research methodologies encompassed \"machine learning\", \"deep learning\", \"neural network\", and \"text mining\". Potential future research hotspots include \"internet of medical things\".</p><p><strong>Conclusion: </strong>This study unveils the current status and development trends of artificial intelligence in chronic disease care, offering novel insights for future artificial intelligence application research.</p>","PeriodicalId":73078,"journal":{"name":"Frontiers in digital health","volume":"7 ","pages":"1608266"},"PeriodicalIF":3.2,"publicationDate":"2025-06-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12224208/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144562238","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Bridging clinic to home: domestic devices in dermatological diagnostics and treatments.","authors":"Diala Haykal, Frederic Flament","doi":"10.3389/fdgth.2025.1595484","DOIUrl":"10.3389/fdgth.2025.1595484","url":null,"abstract":"<p><p>The integration of diagnostic and therapeutic tools into home-used devices has significantly transformed dermatology, making advanced skincare technologies more accessible to the public. Home-based diagnostic devices empower individuals to monitor, assess, and track skin conditions in real time, promoting earlier interventions and personalized skincare. Therapeutic devices, on the other hand, enable users to actively treat cosmetic and dermatological concerns, offering greater autonomy in managing skin health outside the clinical setting. These technologies, often inspired by clinical-grade equipment, promise enhanced patient engagement but also raise critical questions regarding safety, efficacy, and regulatory oversight. Importantly, the regulatory status of these devices, particularly for diagnostic tools, varies significantly across regions, affecting standards for quality, permitted energy outputs, and intended uses. This commentary separately explores the opportunities and challenges posed by home-used diagnostic and therapeutic devices, evaluates their roles in cosmetic dermatology, and highlights key insights from the literature to contextualize their growing influence on personalized skincare.</p>","PeriodicalId":73078,"journal":{"name":"Frontiers in digital health","volume":"7 ","pages":"1595484"},"PeriodicalIF":3.2,"publicationDate":"2025-06-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12213528/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144556076","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Generative AI in healthcare: challenges to patient agency and ethical implications.","authors":"Scott A Holmes, Vanda Faria, Eric A Moulton","doi":"10.3389/fdgth.2025.1524553","DOIUrl":"10.3389/fdgth.2025.1524553","url":null,"abstract":"<p><p>Clinical research is no longer a monopolistic environment wherein patients and participants are the sole voice of information. The introduction and acceleration of AI-based methods in healthcare is creating a complex environment where human-derived data is no longer the sole mechanism through which researchers and clinicians explore and test their hypotheses. The concept of self-agency is intimately tied into this, as generative data does not encompass the same person-lived experiences as human-derived data. The lack of accountability and transparency in recognizing data sources supporting medical and research decisions has the potential to immediately and negatively impact patient care. This commentary considers how self-agency is being confronted by the introduction and proliferation of generative AI, and discusses future directions to improve, rather than undermine AI-fueled healthcare progress.</p>","PeriodicalId":73078,"journal":{"name":"Frontiers in digital health","volume":"7 ","pages":"1524553"},"PeriodicalIF":3.2,"publicationDate":"2025-06-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12213482/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144556077","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}