Frontiers in digital health最新文献

{"title":"Exploring public health equity in digital access: voices from marginalized groups using the \"Mawid\" application in Saudi Arabia.","authors":"Fahad M Al-Anezi","doi":"10.3389/fdgth.2026.1917318","DOIUrl":"https://doi.org/10.3389/fdgth.2026.1917318","url":null,"abstract":"<p><strong>Background: </strong>Digitalization of healthcare services can significantly enhance the services and access to quality care. However, the equity implications of access to healthcare by such digital health platforms for marginalized populations remain underexplored, especially in the countries which are rapidly transforming their healthcare system through digitalization, such as Saudi Arabia. In this context, this study aims to investigate digital inclusion and health equity among users of the Saudi Ministry of Health's \"Mawid\" application.</p><p><strong>Methods: </strong>A qualitative exploratory design was adopted in this study using five focus groups discussions (25 participants with five participants in each group) as an approach for data collection. Each focus group is categorically included participants of older adults, rural residents, low-income individuals, people with disabilities, and intersectional users who combined these characteristics respectively. Braun and Clarke's thematic analysis framework was used to identify key patterns and themes from the data collected through discussions.</p><p><strong>Results: </strong>Findings revealed that Mawid often functioned as a conditional gatekeeper to healthcare, with barriers related to digital literacy, infrastructural constraints, time poverty, and accessibility. Participants relied heavily on informal support networks, frequently at the cost of privacy and autonomy. Uniform system design was perceived to disadvantage users with complex social and health needs.</p><p><strong>Conclusion: </strong>Government-managed digital health applications enhances efficiency of healthcare operations, but may reinforce health inequities, if equity-centred digital design and alternative access pathways are not considered in the system design and formulation of digital health strategies.</p>","PeriodicalId":73078,"journal":{"name":"Frontiers in digital health","volume":"8 ","pages":"1917318"},"PeriodicalIF":5.2,"publicationDate":"2026-09-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13625048/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149119466","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":"Application of artificial intelligence in hepatology.","authors":"Zekai Hu, Yajie Lin, Hainv Gao, Leyao Shuai, Zhaowei Tong, Chenqi Nie, Zhe Yang, Junsheng Zhao","doi":"10.3389/fdgth.2026.1851723","DOIUrl":"https://doi.org/10.3389/fdgth.2026.1851723","url":null,"abstract":"<p><p>Artificial intelligence (AI) is being applied across diagnostic and therapeutic workflows in hepatology. This narrative review summarizes recent advances in AI for liver disease. In medical imaging and digital pathology, computer vision enables automated quantitative analysis of ultrasound, CT, MRI, and histologic images, with the aim of improving the consistency of lesion detection, disease staging, and prognostic assessment. In biomarker research, machine learning can analyze high-dimensional liquid-biopsy and multi-omics data to develop diagnostic and prognostic models; some have outperformed conventional markers in their study cohorts. Electronic health records (EHRs) and large language models (LLMs) are also being investigated for clinical decision support and personalized management. However, most reported evidence remains retrospective, and clinical adoption is limited by data heterogeneity, poor interpretability, uncertain generalizability, and regulatory requirements. Progress will require standardized datasets, external and prospective validation, clinically relevant endpoints, and human-centered implementation before gains in model performance can be translated into better patient outcomes.</p>","PeriodicalId":73078,"journal":{"name":"Frontiers in digital health","volume":"8 ","pages":"1851723"},"PeriodicalIF":5.2,"publicationDate":"2026-09-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13624904/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149119343","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":"CDTD-Agent: a retrieval-augmented multi-agent digital health framework for integrated diabetes care: a single-case expert-rated pilot study.","authors":"Jieyi Zhang, Luqing Wu, Jielong Wu, Xiaoming Han, Lijin Lin, Nengjiang Zhao, Zhihai Zhang, Miaona Cai, Meiqi Chen, Shuyu Yang, Bo Li","doi":"10.3389/fdgth.2026.1923975","DOIUrl":"https://doi.org/10.3389/fdgth.2026.1923975","url":null,"abstract":"<p><strong>Background: </strong>Diabetes management increasingly requires coordination across Western medicine, traditional Chinese medicine (TCM), and lifestyle-based health management. General-purpose large language models (LLMs) can generate clinical text but may lack reliable source grounding and structured cross-disciplinary reasoning.</p><p><strong>Methods: </strong>We developed CDTD-Agent, a retrieval-augmented multi-agent framework that combines a domain-specific knowledge base with an Extract-Retrieve-Reason workflow. The knowledge base included 12 medical textbooks, 8 clinical guidelines, and 183 expert-reviewed diabetes cases. In a blinded single-case pilot evaluation, 30 senior experts rated anonymized recommendations from CDTD-Agent, GPT-4o, DeepSeek-V3.2-Exp, Doubao, and two attending physicians across eight dimensions. Linear mixed-effects models accounted for repeated ratings within evaluators, with Holm correction across eight omnibus tests and global Benjamini-Hochberg correction across 40 focused CDTD-Agent-versus-comparator contrasts. A <i>post hoc</i> backbone-only ablation compared the full framework with Qwen3-235B-A22B using the same case and external prompt without retrieval or role-structured multi-agent processing.</p><p><strong>Results: </strong>Significant source effects were detected in all eight dimensions after Holm correction (all adjusted <i>p</i> < 0.001). After global false discovery rate correction, CDTD-Agent received significantly higher ratings than all three general-purpose LLMs in five dimensions: safety, examination appropriateness, Western medicine treatment appropriateness, TCM treatment appropriateness, and evidence adherence. The largest differences occurred in Western medicine treatment appropriateness, where CDTD-Agent scored 4.07 ± 0.94 versus 1.37-1.60 for the general-purpose LLMs (all adjusted <i>p</i> < 0.001). Non-parametric sensitivity analyses agreed with the mixed-effects results for 39 of 40 focused contrasts. In the <i>post hoc</i> ablation, the full framework outperformed the backbone-only condition in seven of eight dimensions after Holm correction; health management was not significantly different (adjusted <i>p</i> = 0.080).</p><p><strong>Conclusion: </strong>In this standardized complex T2DM case, CDTD-Agent received favorable expert ratings relative to the evaluated general-purpose LLMs in several clinically important dimensions. These findings are scenario-specific and do not establish generalizable clinical performance or isolate the effects of individual framework components. Multicase, repeated-generation, factorial-ablation, external, and prospective validation is needed before clinical deployment.</p>","PeriodicalId":73078,"journal":{"name":"Frontiers in digital health","volume":"8 ","pages":"1923975"},"PeriodicalIF":5.2,"publicationDate":"2026-09-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13623845/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149119430","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":"The epistemic economy of vocal biomarker startups: a qualitative interview study of data practices and governance.","authors":"Abigail Yuhan, Sonal Swain, Robin Zhao, John Ramos, Anaïs Rameau","doi":"10.3389/fdgth.2026.1895383","DOIUrl":"10.3389/fdgth.2026.1895383","url":null,"abstract":"<p><strong>Background and aims: </strong>Vocal biomarker technology has emerged as a promising non-invasive health assessment tool, with artificial intelligence enabling the screening, diagnosis, and monitoring of conditions including Parkinson's disease, Alzheimer's disease, depression, and heart failure. Vocal biomarkers are defined as vocal features, alone or in combination, validated as indicators of clinical outcomes. Startups in this space are generating substantial investment and anticipatory discourse around future commercialization, while the scientific foundation and governance framework remain unestablished. Drawing on Sabina Leonelli's epistemic economy framework, which analyzes how data are produced, valued, and governed within social, institutional, and commercial arrangements, we examine how voice data are converted into entities with clinical and commercial value, and at what scientific and ethical cost.</p><p><strong>Methods: </strong>We performed a qualitative interview study using an interpretive framework. Virtual semi-structured interviews were conducted with representatives from eleven vocal biomarker startups identified through purposive sampling. Interviews addressed data collection methods, IRB compliance, informed consent, participant demographics, geographic sourcing, and financial incentives. Two researchers independently performed thematic analysis, reconciling differences via discussion.</p><p><strong>Results: </strong>Six themes were identified: (1) the absence of shared data standards introduced substantial hardware and protocol heterogeneity, uncontrolled confounding, and label noise, undermining scientific validity; (2) IRB engagement ranged from robust multi-site review to complete absence of oversight; (3) consent practices varied from granular opt-in to implied consent and minimal disclosures; (4) voice data were acknowledged as inherently non-anonymizable, creating unresolved privacy and re-identification risks; (5) data sourcing was asymmetric, with collection from low- and middle-income countries in 4 out of 11 startups while targeting commercial deployment in high-income countries; and (6) commercial incentives produced proprietary protocols that preclude independent validation and impede data sharing.</p><p><strong>Conclusion: </strong>The vocal biomarker startup ecosystem operates as an epistemic economy in which methodological, regulatory, and governance choices tend to favor commercial interests and high-income populations over scientific rigor and equity. Standardized data provenance requirements, reassessment of voice data's risk classification under existing privacy frameworks (e.g., HIPAA, GDPR), and participatory governance including source communities are urgently needed to ensure vocal biomarker products are scientifically valid, ethically grounded, and equitably developed.</p>","PeriodicalId":73078,"journal":{"name":"Frontiers in digital health","volume":"8 ","pages":"1895383"},"PeriodicalIF":5.2,"publicationDate":"2026-09-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13619882/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149083035","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":"Telepsychology and digital mental health communication: an exploratory mixed-methods study of tikTok content.","authors":"Andrea M Carrero, Jobana F Agredo, Lucía Catalina Martínez Ruiz, José A González","doi":"10.3389/fdgth.2026.1900544","DOIUrl":"10.3389/fdgth.2026.1900544","url":null,"abstract":"<p><strong>Background: </strong>Telepsychology consolidated as a mainstream mental health care modality following the COVID-19 pandemic, supported by evidence of clinical efficacy comparable to in-person therapy and a recognized potential to democratize access to care. Against this evolving clinical landscape, social media platforms have simultaneously emerged as major environments for the public circulation of psychological knowledge.</p><p><strong>Objective: </strong>To explore the relationship between scientific evidence on telepsychology and its discursive representation in highly viewed Spanish-language TikTok content, and to identify implications for digital mental health literacy and professional communication practices.</p><p><strong>Methods: </strong>An exploratory mixed-methods design with qualitative emphasis integrated a scoping review (PRISMA-ScR) across Scopus, Web of Science, and PubMed (2020-2025; <i>n</i> = 39 articles) and an inductive thematic analysis of TikTok content (<i>N</i> = 50 videos; 1,286,360 total views). Searches were conducted from a newly created TikTok account with no prior interaction history, configured in Colombia, to reduce algorithmic personalization bias. Three exploratory corpus-descriptive indicators-the Telepsychology Gap Index (TGI), Digital Telepsychology Legitimacy Index (DTLI), and Simplified Circulation Index (SCI)-were computed as heuristic comparators.</p><p><strong>Results: </strong>The scientific literature consistently supported the clinical viability and democratizing potential of telepsychology. Within the TikTok corpus, 64% of content featured simplified definitions or subjectivist experiential narratives with minimal reference to ethical or regulatory frameworks. Corpus-descriptive indicators revealed a substantial gap between scientific representation and digital circulation (TGI = 0.80), limited presence of clinical foundations in digital discourse (DTLI = 0.36), and a predominance of simplified over technically grounded content (SCI = 1.78).</p><p><strong>Conclusions: </strong>Within the analyzed corpus, telepsychology undergoes a process of digital discursive reinterpretation in which communicative accessibility and immediate subjective experience displace clinical precision as central evaluative criteria. These findings underscore the need to strengthen digital mental health literacy and professionals' evidence-based communication skills online. Given the exploratory design, findings are specific to a non-probabilistic Spanish-language TikTok corpus and are not generalizable prevalence estimates. Future research should extend this agenda through larger multilingual samples, additional platforms, computational social science approaches, and longitudinal designs examining how digital representations of telepsychology shape users' expectations and engagement with digital mental health care.</p>","PeriodicalId":73078,"journal":{"name":"Frontiers in digital health","volume":"8 ","pages":"1900544"},"PeriodicalIF":5.2,"publicationDate":"2026-09-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13621891/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149083076","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":"Impact of coping profiles on social identity in a youth chronic pain recommender system: qualitative descriptive study.","authors":"Antonia Barbaric, Joseph Saliba, Anya Nair, Vina Mohabir, Jennifer Stinson, Alex Mariakakis, Quynh Pham, Joseph A Cafazzo, Chitra Lalloo","doi":"10.3389/fdgth.2026.1727940","DOIUrl":"10.3389/fdgth.2026.1727940","url":null,"abstract":"<p><strong>Background: </strong>Chronic pain affects one in five children and adolescents worldwide yet remains one of the most underrecognized and undertreated illnesses, emphasizing the need for accessible, personalized digital health support. The Power over Pain portal delivers such support through a virtual stepped-care pathway, but like many digital interventions, is vulnerable to declining user engagement over time. Health recommender systems (HRSs) can personalize resource delivery and combat \"choice overload,\" thereby encouraging sustained use. This study investigates the application of a health recommender system (HRS) as a way of increasing engagement with relevant virtual care resources for pediatric chronic pain. The perceived value of the HRS was evaluated through a qualitative descriptive study involving ten participants diagnosed with chronic pain.</p><p><strong>Results: </strong>Thematic analysis revealed four main themes: (1) HRS serves as a catalyst for important dialogue in routine care; (2) coping results from the HRS are selectively applied by users according to their needs; (3) the HRS creates a personable experience, positively influencing willingness to engage in care; (4) interacting with the HRS may affirm participants' social identities, fostering a sense of belonging in the pain community and potentially enhancing long-term resource engagement.</p><p><strong>Conclusion: </strong>These findings challenge the traditional focus on recommendation accuracy alone, suggesting instead that the effectiveness of an HRS may also be influenced by design features that support validation, shared experience, and potential social identification, which can deepen engagement and promote involvement in care. Future work should investigate the impact of mechanisms that reinforce users' social identities to understand their effects on sustained use and self-management behaviors.</p>","PeriodicalId":73078,"journal":{"name":"Frontiers in digital health","volume":"8 ","pages":"1727940"},"PeriodicalIF":5.2,"publicationDate":"2026-09-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13619948/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149083131","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":"Decoding digital transformation in health professions education: a comparative analysis of knowledge, attitudes, and readiness among faculty and students at KSAU-HS.","authors":"Amal I Khalil, Lamyaa Y ALyaba, Amnah M Jambi","doi":"10.3389/fdgth.2026.1881377","DOIUrl":"10.3389/fdgth.2026.1881377","url":null,"abstract":"<p><strong>Introduction: </strong>Digital transformation is altering educational processes in health professions and needs instructors and students to acquire competencies for its implementation. At the same time, comparative information about readiness for digital transformation for educators and students in Saudi Arabian universities is scarce and the reasons for such readiness are not fully known. This study reviewed the knowledge, attitude and readiness of instructors and learners at King Saud bin Abdulaziz University for Health Sciences.</p><p><strong>Methods: </strong>The study used a cross-sectional comparative design. Data was collected from the respondents with the help of a specially designed questionnaire, based on the Technology Acceptance Model. The questionnaire measured digital competency, perceived usefulness and ease of use and facilitating conditions.</p><p><strong>Results: </strong>Respondents displayed a slightly more developed level of digital competence, perceived usefulness, perceived ease of use, and readiness for digital transformation. As for students, motivation and confidence in one's ability to work with digital technologies turned out to be the best predictors. Structural equation modeling has validated that usefulness and ease of use along with facilitating conditions represent significant factors for predicting the intention to use digital technologies. Hence, psychology and organizational aspects play a bigger role in being ready for the digital transformation than demographic variables.</p><p><strong>Discussion: </strong>The readiness for digital transformation in health professions education relates to not only the technical aspect but also to individual motivation, confidence, and institutional backing. Therefore, universities should develop specific strategies to motivate and encourage students to learn, along with training faculty, giving sufficient resources, and offering suitable organizational conditions.</p>","PeriodicalId":73078,"journal":{"name":"Frontiers in digital health","volume":"8 ","pages":"1881377"},"PeriodicalIF":5.2,"publicationDate":"2026-09-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13619956/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149083060","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":"From data to action: the FRAME conceptual framework for translating consumer wearable feedback into behavior change support.","authors":"Sundus Malaikah","doi":"10.3389/fdgth.2026.1871670","DOIUrl":"10.3389/fdgth.2026.1871670","url":null,"abstract":"<p><strong>Background: </strong>Consumer wearables are now widely used across health promotion, wellness, and recreational fitness. Although wearable-supported interventions can improve physical activity and related outcomes, effects are typically modest, heterogeneous, and more consistent when devices are embedded within broader behavior change support rather than used alone. In applied settings, however, wearable use often remains focused on metric display and passive monitoring, leaving practitioners without a clear framework for translating data into behaviorally meaningful support.</p><p><strong>Objective: </strong>This paper proposes FRAME, a theory-informed conceptual framework intended to help practitioners interpret consumer wearable outputs as behaviorally relevant signals rather than as self-sufficient indicators of progress or failure.</p><p><strong>Methods: </strong>The framework was developed through a targeted narrative synthesis of literature on wearable-supported interventions, self-monitoring, feedback design, engagement, adaptive support, and applied behavior change theory. The purpose of the synthesis was not to estimate pooled effects, but to identify recurring interpretive and implementation problems that arise when wearable data are used in real-world behavior support encounters.</p><p><strong>Framework overview: </strong>FRAME comprises five linked steps: Filter the signal, Read the behavior pattern, Align a behavior change target, Message and micro-plan, and Evaluate and iterate. Across these steps, wearable data are treated as inputs into interpretation, hypothesis generation, collaborative planning, and iterative review. The framework is grounded primarily in self-regulation, the Capability-Opportunity-Motivation-Behavior (COM-B) model, self-determination theory, social cognitive theory, habit formation, and relapse prevention.</p><p><strong>Conclusion: </strong>For practitioners, the framework offers a structured way to judge whether a wearable signal is usable, what behavior pattern it may reflect, which target to prioritize, and how to discuss the data in an autonomy-supportive way. For wearable users, this is intended to yield more collaborative and sustainable behavior change support.</p>","PeriodicalId":73078,"journal":{"name":"Frontiers in digital health","volume":"8 ","pages":"1871670"},"PeriodicalIF":5.2,"publicationDate":"2026-09-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13619601/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149083070","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":"Growing public interest and regional disparities of cardiorenal syndrome in the United States.","authors":"Reid P Schlesinger, Kianoush B Kashani, Charat Thongprayoon, Wisit Cheungpasitporn, Jing Miao","doi":"10.3389/fdgth.2026.1803510","DOIUrl":"10.3389/fdgth.2026.1803510","url":null,"abstract":"<p><strong>Background: </strong>Public awareness and regional engagement with cardiorenal syndrome (CRS) remain poorly understood. We examined temporal trends and geographic patterns of U.S. public interest in CRS and their relationship with disease burden.</p><p><strong>Methods: </strong>Google Trends™ data (2004-2025) annual relative search interest (RSI) for CRS. CRS-related mortality was derived from CDC WONDER (2007-2020) using hypertensive heart and chronic kidney disease deaths. Temporal trends were assessed via augmented Dickey-Fuller tests and linear regression. Associations between RSI and mortality were evaluated using Pearson correlation, AR(1) modeling, first-difference, and cumulative change analysis. State-level correlations were examined with and without adjustment for income, education, and insurance coverage.</p><p><strong>Results: </strong>Both RSI and mortality exhibited strong non-stationary upward trend (<i>p</i> = 0.98 and 0.42). Raw data showed a strong correlation analysis (r = 0.79, <i>p</i> = 0.001); but first-differenced year-over-year changes were not significant (r = -0.36, <i>p</i> = 0.23), consistent with AR(1) modeling (<i>β</i>=1.58, <i>p</i> = 0.11). Cumulative changes remained strongly correlated (r = 0.80, <i>p</i> = 0.001), indicating parallel long-term trajectories. Findings were consistent with varying ICD-10 code definitions for CRS-related mortality. Geographically, RSI was highest in West Virginia and lowest in Texas and Georgia. State-level RSI showed no correlation with age-adjusted mortality, both unadjusted (r = -0.09, <i>p</i> = 0.52) and after adjustment for median income, educational, and insurance coverage (<i>β</i>=0.03, <i>p</i> = 0.75).</p><p><strong>Conclusion: </strong>Public interest in CRS has grown in parallel with mortality, reflecting its increasing epidemiological relevance. However, the lack of geographic correlation between search interest and mortality-even after socioeconomic adjustment-suggests awareness gaps and highlights opportunities for targeted education and resource allocation.</p>","PeriodicalId":73078,"journal":{"name":"Frontiers in digital health","volume":"8 ","pages":"1803510"},"PeriodicalIF":5.2,"publicationDate":"2026-09-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13619914/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149083116","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":"Lightweight structure-aware explanations for rule-based models: application to radiotherapy risk communication in breast cancer.","authors":"Alessandro Umbrico, Guido Bologna, Luca Coraci, Francesca Fracasso, Silvia Gola, Gabriella Cortellessa","doi":"10.3389/fdgth.2026.1925421","DOIUrl":"10.3389/fdgth.2026.1925421","url":null,"abstract":"<p><strong>Objective: </strong>Rule-based models are often considered inherently interpretable because their explicit logical structure allows users to trace predictions back to human-readable rules. However, in complex domains such as healthcare, multiple rules involving numerous attributes may be activated simultaneously, making it difficult for end users to identify the most influential factors behind a prediction. This can limit practical transparency, particularly in clinical shared decision-making, where clear and trustworthy explanations are essential. An example is communicating the risk of side effects following radiotherapy for breast cancer, where understanding the determinants of estimated risk may support informed treatment choices.</p><p><strong>Method: </strong>We propose a lightweight, structure-aware attribution method for rule-based models inspired by information retrieval techniques. The approach represents the activated rules associated with a prediction as a document-like structure and applies a TF-IDF-inspired weighting scheme to estimate the relevance of individual input attributes. Unlike perturbation-based methods such as SHAP and LIME, the proposed approach operates directly on the internal model structure, without requiring input perturbations, surrogate models, repeated evaluations, or access to training data. This preserves the explanatory signal encoded in the rules while reducing computational complexity and execution time. The method is evaluated through a technical comparison with SHAP and LIME and a user-centered study involving potential end users.</p><p><strong>Results and conclusions: </strong>Experimental results show moderate agreement between the TF-IDF-inspired ranking and established attribution methods, identifying attributes that are behaviorally informative with respect to the implemented predictor. Faithfulness analysis shows that direct rule frequency achieves higher perturbation faithfulness, highlighting a trade-off introduced by the proposed weighting scheme. Complementarily, the user study indicates that the complete explanation pipeline, combining relevance weighting, semantic aggregation, and user-oriented presentation, is perceived as more interpretable and useful than raw rule-based explanations. These findings suggest that information-retrieval-inspired weighting can provide a lightweight mechanism for organizing rule evidence, while semantic abstraction and user-oriented presentation can facilitate its communication to end users.</p>","PeriodicalId":73078,"journal":{"name":"Frontiers in digital health","volume":"8 ","pages":"1925421"},"PeriodicalIF":5.2,"publicationDate":"2026-09-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13617116/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149076781","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}