Healthcare

Generative AI tools such as ChatGPT are increasingly used by the public to seek guidance on diet and physical activity for type 2 diabetes (T2D) prevention and management. However, the consistency of model outputs across different users and disease-stage scenarios remains insufficiently characterized. This pilot study aims to evaluate the word-level and semantic-level consistency of GPT-4os diet and physical activity responses for type 2 diabetes prevention and management. We designed 12 prompts covering four categories: prediabetes, diagnosed type 2 diabetes (T2D), diagnosed T2D with complications, and general questions that did not specify dysglycemia stage. Word-level similarity was quantified with Term Frequency-Inverse Document Frequency (TF-IDF) cosine scores; sentence-level semantic similarity was measured using large language models (LLMs) - DeBERTa-v3 MNLI to calculate the entailment probabilities. The results showed that mean cosine similarity across users was moderate (0.44-0.66), whereas mean entailment similarity was higher (0.68-0.81). Across stages, word-level similarity was low to moderate (0.28-0.63) and entailment similarity remained moderate to high (0.63-0.80). Low similarity commonly referenced distinct food choices, operational details, safety warnings, and stage-specific suggestions. GPT-4o generated semantically consistent but variably detailed responses and the moderate semantic variation suggested limited differentiation of response content across diabetes-related stages in this pilot consistency assessment.

Background: The growing burden of lifestyle-related chronic diseases has increased the need for clinically interpretable decision-support tools capable of integrating artificial intelligence with evidence-based preventive nutrition. Although machine learning has shown considerable potential for health risk prediction, most existing approaches remain limited to isolated predictive models or conventional nutritional software, with little integration of multidimensional clinical assessment and personalized recommendations. Objective: To develop and internally validate NutrIA, a hybrid web-based Clinical Decision Support System (CDSS) that combines machine learning, validated clinical assessment, structured clinical reasoning and personalized nutritional recommendations for preventive medicine. Methods: NutrIA was developed using harmonized data from the National Health and Nutrition Examination Survey (NHANES, 1988 to 2018). A supervised machine learning model was trained to estimate 5-, 10- and 20-year all-cause mortality risk and subsequently integrated with an adaptive clinical questionnaire, validated screening instruments, nutritional indicators, dietary clustering, clinical phenotyping and a transparent rule-based recommendation engine within a unified web-based platform. Results: The predictive model achieved ROC-AUC values of 0.894, 0.914 and 0.923 for 5-, 10- and 20-year mortality prediction, respectively. The implemented CDSS incorporates an adaptive questionnaire (151 items), 39 validated clinical assessment instruments, 17 clinical phenotypes and 31 dietary clustering modules to generate individualized nutritional and lifestyle recommendations together with an automated clinical report. The integrated framework translates probabilistic risk estimates into clinically interpretable decision support for personalized preventive nutrition. Conclusions: NutrIA demonstrates the technical feasibility of integrating machine learning with knowledge-based clinical reasoning within a single web-based CDSS for preventive nutrition. Although external validation and prospective clinical evaluation are required before routine implementation, the proposed architecture represents a promising step toward clinically interpretable artificial intelligence for personalized nutritional care.

Background: Reducing maternal mortality and improving newborn and child outcomes are targets of the Sustainable Development Goals. Evidence on how these efforts are reaching women with disabilities is lacking. Objectives: to estimate global, relative inequalities in stillbirth, neonatal, infant, and maternal mortality for women with disabilities compared to women without disabilities. Methods: We searched MEDLINE, Global Health, PsycINFO, and Embase from 1 January 2015 to 28 January 2026, to identify articles on disability and stillbirth, neonatal, infant, and maternal mortality. We included studies that had a recognised measure of disability as an exposure, a control group of women without disabilities and at least one of the four outcomes. A pooled estimate for each outcome was done using a random-effects meta-analysis of the minimally adjusted results. Results: We identified over 4,300 titles, of which 17 papers were eligible for inclusion. Almost all data came from nationally-representative data sources in high-income countries. We found that women with disabilities were 4.66 times more likely (95% C.I. 1.70-12.75) to experience maternal mortality compared to women without disabilities. Women with disabilities were also 37% more likely to have a stillbirth and 27% and 48% more likely to have a neonatal or infant death compared to women without disabilities, respectively. Conclusions: Women with disabilities consistently have higher incidence of maternal, stillbirth, neonatal, and infant mortality, even in countries that have relatively low incidence of these outcomes. There is a lack of evidence globally and particularly from LMICs, and on effective interventions to improve maternal and infant outcomes for women with disabilities.

Against the backdrop of nationwide inclusive education promotion, children with autism spectrum disorder, intellectual disabilities and other special educational needs (SEN) in Jiangxi Province have raised growing demands for equitable schooling. Paraeducators serve as a critical on-site support mechanism enabling SEN children to access mainstream classrooms; the adequacy of shadow teacher service provision and the maturity of corresponding multi-stakeholder support systems jointly determine the overall quality of local inclusive education. This study adopted mixed quantitative-qualified methods, including questionnaire surveys and semi-structured interviews, to investigate SEN children, paraeducators, general and special education teachers, as well as SEN caregivers across multiple prefecture-level cities in Jiangxi. Grounded in provincial special education policies and local frontline inclusive education practices, we systematically unpacked multidimensional service demands from four core stakeholder groups, diagnosed prominent practical bottlenecks restricting the sustainable operation of shadow teacher services, and constructed regionally tailored multi-layered support strategies aligned with Jiangxi's educational realities. The findings of this research offer empirical evidence and actionable policy references to advance high-quality inclusive education for SEN children across central China's Jiangxi Province.

Aim: Little is known about patient safety in a less consolidated obstetric system where various facilities, such as perinatal medical centers (PMCs), general hospitals, and clinics, collaborate under risk-based role differentiation. We aimed to compare maternal complications after cesarean section by facility type across area types (rural, provincial, and metropolitan) in Hokkaido, Japan. Methods: This retrospective cohort study used insurance claims data from Hokkaido (2018-2025). Two comparisons were conducted for a composite outcome of postpartum hemorrhage, infection, and thrombosis: a two-category comparison (PMC vs. non-PMC, combining general hospitals and clinics) across all areas, with an interaction term between facility and area type; a three-category comparison (PMC vs. general hospital vs. clinic) restricted to metropolitan areas. Generalized estimating equations with a Poisson distribution, accounting for clustering within facilities, were applied to estimate risk ratios. Results: A total of 1,822 participants underwent cesarean section. PMCs were associated with lower maternal complication rates compared to non-PMC facilities (adjusted RR 0.37, 95% CI 0.16-0.88 in rural areas; adjusted RR 0.20, 95% CI 0.11-0.37 in provincial areas). In metropolitan areas, PMCs and general hospitals were associated with lower maternal complication rates compared to clinics (PMC vs clinic: adjusted RR 0.42, 95% CI 0.18-0.97; general hospital vs clinic: adjusted RR 0.25, 95% CI 0.09-0.70). Conclusions: Higher-level facilities were associated with lower maternal complication rates after cesarean section in Japan. These findings provide important evidence for regional consolidation of obstetric care.

Introduction: Understanding of absolute and relative estimates (i.e., effect size), and certainty of evidence corresponding to those estimates, is a fundamental evidence-based practice competency to promote informed clinical decision-making. While research has been conducted in the medical profession, there is no published research on these competencies in the nutrition and dietetics profession. Methods: Among registered dietitians, our main objectives were to assess (1) their understanding and perceived usefulness of three absolute and two relative estimate approaches to assess effect size, (2) their perceived usefulness of certainty of evidence, and (3) factors influencing their understanding and perceived usefulness. We conducted a web-based, cross-sectional survey among dietitians recruited from the Academy of Nutrition and Dietetics (United States). Participants received effect estimates based on hypothetical dietary interventions vs. usual diet for reducing myocardial infarction risk. Results: Of the 11,050 dietitians who received the survey link, 210 participated (2.0% response rate), and only completers (n=114) were included in the analysis. Participants demonstrated a similar understanding of the relative (27.6%) and absolute (27.5%) estimates, with Risk Difference (30.7% correct responses) being the best understood approach and Number Needed to Treat (24.6%) being the least. The understanding of five approaches was not different than random guessing (p>0.05). While perceived usefulness scores were similar between five approaches, they were highest when data was presented as Relative Risk [mean (SD): 4.82 (1.50)]. Dietitians rated the usefulness of certainty of evidence favorably [mean (SD): 5.07 (1.83), on a 7-point scale), and no factors were associated with correct understanding. Conclusion: Dietitians may have limited understanding of how to interpret effect sizes, a finding consistent with surveys of other health professionals. To optimize informed decision-making between dietitians and clients, dietetic programs and continuing education platforms should consider additional training on interpreting effect sizes and certainty of evidence for effect sizes.

Background: Whether corrective actions documented in medical safety incident reports rely on individual vigilance ("Safety-I") or on structural, system-level intervention ("Safety-II") has not been quantitatively evaluated on a national scale in Japan. We developed an automated classification pipeline to assign corrective-action free-text to a 7-level maturity scale (L0-L6) and computed two summary indices: the Safety Measure Quality Profile (SMQP), the full L0-L6 distribution, and the System-based Safety Measure Rate (SSMR), the proportion of non-L0 records classified L3-L6. Methods: We analyzed all 11,507 corrective-action free-text entries from the 2010 release of Japan's national medical accident and near-miss reporting database (Japan Council for Quality Health Care, JCQHC), comprising 8,804 near-miss (Hiyari-Hatto) and 2,703 accident (Jiko) reports. Records were classified using a five-stage hybrid pipeline: an expert-developed rule dictionary, TF-IDF + k-nearest-neighbor matching, cosine-similarity matching, a two-tier large-language-model (LLM) classifier, and a conservative priority-cascade fallback. SSMR was compared between near-miss and accident reports using a chi-square test, Wilson 95% confidence intervals, Cramer's V, and the risk difference (RD), against pre-specified minimal clinically important difference (MCID) criteria of RD >= 2 percentage points and Cramer's V >= 0.10. Results: Every record received a definitive L0-L6 label (0% unresolved). Overall, 16.6% of records were unclassifiable (L0); among the 9,599 classifiable (non-L0) records, individual-vigilance actions (L1) predominated (54.6% of all records), and only 11.82% (95% CI, 11.19-12.49%) met the SSMR criterion (L3-L6). SSMR was higher for accident reports than for near-miss reports (18.12% [95% CI, 16.69-19.64%] vs. 9.[95% CI,46% [95% CI, 8.80-10.17%]; RD = 8.66 percentage points; Cramer's V = 0.120; chi-square(1) = 136.97001), exceeding both pre-specified MCID thresholds. Conclusions: In this interim single-year analysis, the large majority of documented corrective actions in Japanese medical safety reports remained individual-vigilance-based rather than system-based, with accident reports showing a substantively, rather than merely statistically, higher proportion of system-based actions than near-miss reports. These findings support the feasibility of large-scale automated assessment of corrective-action quality and provide the rationale for the planned 16-year longitudinal analysis.

Background: Stroke is among the foremost contributors to mortality and lasting disability and continues to place a heavy clinical and societal burden worldwide. Although metabolic syndrome (MetS) is recognized as a contributor to stroke risk, insufficient attention has been paid to how this relationship behaves when potential confounders are entered into the model in a stepwise manner. Objectives: This study aimed to characterize the relationship between MetS and stroke prevalence using a series of sequentially adjusted models built from the Korea National Health and Nutrition Examination Survey (KNHANES). We explored whether self-rated health is a useful functional indicator for stratifying stroke risk. Methods: Of the 22,559 KNHANES VIII respondents, 12,536 participants aged [≥]19 years and with information required to classify physician-diagnosed stroke (DI3_dg) or define MetS were included in the final analysis. Associations were estimated using complex-sample logistic regression under a sequential adjustment scheme: Model 1 (unadjusted), Model 2 (adjusted for age and sex), Model 3 (further adjusted for educational level and family history of stroke), and Model 4 (additionally incorporating economic activity status and self-rated health). Results: MetS was associated with an increased risk of stroke in all models: Model 1 (odds ratio [OR] 3.372, 95% confidence interval [CI] 2.443-4.654), Model 2 (OR 1.956, 95% CI 1.396-2.740), Model 3 (OR 1.813, 95% CI 1.292-2.546), and Model 4 (OR 1.636, 95% CI 1.153-2.321). A graded pattern was noted concerning self-rated health, with progressively poorer perceived health corresponding to higher odds of stroke, and the "very poor" category showed substantially elevated odds (OR 9.836 in Model 4). Conclusions: MetS was independently associated with stroke prevalence even after sequential adjustment. Self-rated health appears to capture both metabolic burden and broader functional health aspects.

Cancer prevalence is distributed unevenly across regions and caused by the interaction of multiple risk factors. Previous studies focused on the use of global modeling techniques to predict cancer at the county level that overlooks important spatial differences. This study aims to develop geographically weighted machine learning models to predict cancer prevalence at the census tract level in the United States and identify local determinants of cancer burden. First, a scoping review was conducted to find a list of measurable drivers of cancer in the United States. Using this list, the data of these variables for 84415 census tracts were obtained from the Center for Disease Control and Prevention PLACES dataset and other publicly accessible resources. Then, several predictive models, including Ordinary Least Squares (OLS) and Geographically Weighted Regression (GWR), as well as Random Forest, XGBoost, and Deep Neural Network and their geographically weighted counterparts, were developed and compared using the Coefficient of Determination, Root Mean Square Error, and Absolute Error. Results presented that geographically weighted models outperformed other methods, and geographically weighted XGBoost achieved the strongest and most consistent overall performance with pseudo-R2 ranging between 0.89 and 0.98. Feature importance analysis of this model illustrated that most important cancer drivers changed location by location. Aged people, racial composition, preventative behaviors, and metabolic conditions such as diabetes, hypertension, and high cholesterol were determined as influential predictors, although their relative importance varied across regions. These findings revealed the value of localized models at a small geographic scale to identify regional cancer risk patterns and help the allocation of proper resources to hotspot areas. Keywords: Cancer prevalence, Census tracts, geographically weighted machine learning models, Deep neural network, XGBoost, Random Forest, Ordinary Least Squares, risk factor, determinant

Abstract Objective To evaluate the adherence to guidelines and readability of large language model-generated sexual health information related to female sexual dysfunction following cystectomy, and to determine whether adherence differs across models and prompt formats. A secondary objective was to introduce an analytic strategy using principal component analysis to examine the dimensions of readability metrics. Methods Three large language models (LLMs), ChatGPT, Gemini, and Perplexity were prompted with six clinical questions related to sexual function after cystectomy. Questions were phrased in long-form and short-form language. Responses were independently graded by two reviewers, derived from guideline recommendations. Linear mixed-effects models predicted adherence as functions of LLM, prompt, and reviewer, with clinical questions as a random intercept. Readability was assessed using five metrics, and principal component analysis (PCA) was used to determine latent structure. Results ChatGPT demonstrated the highest (estimated marginal mean [emm] = 0.769), outperforming Gemini (0.499) and Perplexity (0.457). Shorter, less complex prompts elicited higher adherence than more complex, clinical prompts. All models produced content that exceeded recommended reading levels. PCA demonstrated that a single dominant component accounted for 76.7% of variance across readability indices, indicating a shared underlying construct. Conclusion ChatGPT produced the most guideline-concordant information overall. High linguistic complexity was seen across models, highlighting a barrier to patient comprehension. These findings characterize large language models as variable medical information systems whose outputs rely heavily on prompt structure and model type.

Introduction: The advent of Endovascular surgery and the use of fluoroscopy-guided procedures have grown in the last decade, and with that increased the exposure of surgeons to cumulative ionizing radiation, increasing the risk of occupational health complications. Despite established radiation protection principles, adherence to radioprotection measures and surveillance practices remains uncertain in many settings. This study aimed to evaluate knowledge, availability, and implementation of radiation protection strategies among vascular surgeons in Brazil Metology: A national cross-sectional survey was conducted in February 2026 using an anonymous online questionnaire distributed to all active members of the Brazilian Society of Angiology and Vascular Surgery (SBACV). Associations between participant characteristics and radioprotection practices were explored using chi-square or Fishers exact tests. Results: Of 4,698 invited members, 192 vascular surgeons met the inclusion criteria. Most participants were male (67.4%), with a mean age of 45.9 years and a median of 12 years of experience performing fluoroscopy-guided procedures. Basic PPE use, particularly lead aprons, was nearly universal; however, adherence to other protective measures was substantially lower. Most respondents reported employing radiation-reduction strategies, including minimizing fluoroscopy time (98.9%), collimation (90.6%), optimized table and detector positioning (88.3%), procedural planning (87.2%), and pulsed fluoroscopy (81.0%). Only 19.4% reported undergoing annual medical surveillance for radiation-related health effects. Cataracts were the most frequently reported radiation-associated condition (6.2%). Greater age and professional experience were associated with higher utilization of selected protective measures and advanced imaging strategies. Conclusion: The study provides important insights into how radiation protection is currently understood and implemented in contemporary Brazilian vascular surgery practice. The incomplete use of personal protective equipment (PPE), with many justifications in addition to the low surveillance of the effects of radiation on health, brings us an alert about the reality in Brazil.

Wayfinding in hospitals is often hindered by ineffective signage; however, the cognitive mechanisms of healthcare wayfinding symbols comprehension remain under-researched. This study utilized eye-tracking and spatial gaze mapping to examine how visual complexity, abstraction, and human figuration modulate perception in 40 healthy adults viewing 24 hospital-related healthcare wayfinding symbols. Results indicate that pupil size is a sensitive physiological marker of cognitive load, significantly influenced by visual complexity ({chi}2 = 11.32, p = .022) and abstraction ({chi}2 = 7.49, p = .027). Human figuration reduced fixation duration and increased saccade amplitude, facilitating efficient semantic integration. Furthermore, human-centric healthcare wayfinding symbols elicited streamlined gaze trajectories, whereas abstract/complex designs induced chaotic scanpaths. These findings suggest that human figuration acts as a cognitive scaffold, reducing mental effort. We provide evidence-based guidelines for optimizing healthcare wayfinding symbols by prioritizing human body representations and balancing abstraction levels. HighlightO_LIPupil size indexes cognitive load during symbol comprehension. C_LIO_LIHuman figuration cuts fixation duration, boosting wayfinding efficiency. C_LIO_LIAbstract symbols increase pupil dilation, raising cognitive load. C_LI

Background: University students experience substantial psychological well-being and body-image concerns, while scalable, personalized digital support remains underexamined in Malaysia. Artificial intelligence chatbots may deliver repeated lifestyle guidance, but the incremental value of personalization over structured chatbot support is uncertain. Objectives: This study evaluated changes in psychological well-being and body appreciation following a 12 week personalized AI-powered lifestyle intervention, NExGEN, among Malaysian university students. Methods: A two-arm, controlled, quasi-experimental pre-post study allocated 140 students aged 18 to 35 years by matched blocks to NExGEN (n = 70) or a structured-prompt ChatGPT control (n = 70). NExGEN generated adaptive weekly lifestyle actions from a 47-item onboarding assessment, whereas control participants received standardized weekly prompts covering the same lifestyle domains. Psychological well-being and body appreciation were assessed at baseline and week 12 using the World Health Organization-Five Well-Being Index and Body Appreciation Scale-2. Intention-to-treat linear mixed models estimated adjusted within-group changes and between-group differences in change, with Holm adjustment for the co-primary outcomes. Results: Week-12 assessments were completed by 121 participants (86.43%). In NExGEN, psychological well-being improved by an adjusted 8.68 points (95% CI, 6.22 to 11.14), z = 6.91, p < .001, and body appreciation improved by 0.17 points (95% CI, 0.10 to 0.24), z = 4.82, p < .001. However, between-group differences in change were not statistically significant for psychological well-being (2.87 points; 95% CI, -0.48 to 6.23; z = 1.68; Holm-adjusted p = .093) or body appreciation (0.10 points; 95% CI, 0.00 to 0.19; z = 1.99; Holm-adjusted p = .093). Median platform logins were 68.00 in NExGEN and 58.50 in control; mean acceptability scores were 3.92 and 3.59, respectively. Conclusions: NExGEN participation was associated with significant within-group improvements in psychological well-being and body appreciation, but personalized guidance did not demonstrate superiority over structured chatbot guidance. Because allocation was quasi-experimental, causal attribution remains limited. Randomized component-level trials are needed to determine whether personalization provides incremental benefit.

Objectives: Dentists prescribe approximately one in ten antibiotics worldwide, yet antimicrobial stewardship (AMS) remains underemphasized in dental education. Large language models (LLMs) may support AMS training, but their proficiency and clinical reasoning in this context remain unclear. We evaluated GPT-4o's accuracy and clinical reasoning on dental antibiotic prescribing questions, stratified by question difficulty. Methods: We assembled 125 multiple-choice questions on dental antibiotic prescribing from eight peer-reviewed studies (2017-2023). GPT-4o answered each question and generated a clinical justification. Accuracy was assessed against source-study answer keys and examined across difficulty quartiles. Justifications were evaluated using an adapted 12-axis human-evaluation framework assessing scientific consensus, extent and likelihood of harm, inappropriate and missing content, bias, and both correct and incorrect comprehension, retrieval, and reasoning. Prophylaxis-specific questions were analysed separately. Results: GPT-4o correctly answered 72% of questions. Accuracy remained relatively stable across difficulty quartiles (78%, 78%, 65%, 70%). Experts rated 95.4% of justifications positively across the 12 axes. Comprehension, retrieval, and reasoning each exceeded 96.2% positive ratings. Missing content was the main weakness (7.8%), and 7.1% of justifications showed a moderate-to-severe potential for harm. Performance on prophylaxis-specific questions (98.1%) exceeded non-prophylaxis questions (93.0%). Conclusions: GPT-4o demonstrated moderate-to-high proficiency and clinically defensible reasoning in dental antibiotic prescribing questions. However, residual risks indicate that it is not suitable for unsupervised clinical use but shows potential as a supervised AMS educational tool.

Abstract Background: Stroke is a neurological disorder which is defined as the sudden onset of focused or global disruptions in functions of brain caused due to vascular issue which lasts more than 24 hours or sometimes leading to death. Objective: To determine the effects of multimodal balance training with and without auditory cues on balance, gait mobility, risk of fall and quality of life in patients with chronic stroke. Methodology: This randomized controlled trial, conducted at Islam Teaching Hospital and Idrees Hospital Cant. Sialkot, Pakistan, included 21 stroke survivors per group, 42 in total, (aged 45-70, 1-year post-stroke) using non-probability convenient sampling. Group A received multimodal balance training with auditory cues, while Group B received the same training without cues for 12 weeks. Exclusion criteria included respiratory or orthopedic conditions, cognitive disorders (MMSE < 24), aphasia, non-healing ulcers, or osteoporosis. Outcomes (Berg Balance Scale, Time Up and Go Test, Fall Efficacy Scale-International, Stroke Specific Quality of Life Scale) were assessed at baseline, 6 weeks, and 12 weeks. Results: Group A (with auditory cues) showed statistically significant improvements in balance (Berg Balance Scale: median 21 to 47.5, p < .001), gait mobility (Time Up and Go Test: median 26 to 11 seconds, p < .001), fall risk (Fall Efficacy Scale-International: median 61 to 17, p < .001), and quality of life (Stroke Specific Quality of Life Scale: median 91 to 176.5, p < .001) over 12 weeks, outperforming Group B (without auditory cues) in all measures (p < .001 for balance, gait, and fall risk; p = 0.001 and p < .001 for quality of life at 6 and 12 weeks, respectively). Conclusion: Chronic stroke treatments including multimodal balance training with auditory cues have demonstrated significant advantages over a 12-week therapy session. The results demonstrate significant improvements in balance, gait mobility, risk of fall, and quality of life in chronic stroke survivors. Abbreviations: MMBT (Multimodal Balance Training), MMBTwAC (Multimodal Balance Training with Auditory Cues, referring to Group A), RAS (Rhythmic Auditory Stimulation), RCT (Randomized Controlled Trial), MMSE (Mini-Mental State Examination), BBS (Berg Balance Scale), TUG (Time Up and Go Test), FES-I (Fall Efficacy Scale-International), SSQOL (Stroke-Specific Quality of Life Scale), SPSS (Statistical Package for the Social Sciences), SD (Standard Deviation), and MAS (Modified Ashworth Scale). Key words: Multimodal, Balance, Stroke, Gait, Berg Balance Scale (BBS) and Auditory cues.

There were substantial changes in NIH policies regarding research funding in FY2025. This work examines NIH funding for pediatric research FY2020 through Q2FY2026 including the number and cost of awards, the number of first year (type 1) awards, the number of Notices of Funding Opportunity, and the topical focus of research awards. NIH funding for pediatric research declined >20% in the first two quarters FY2025-FY2026 with proportionally greater reductions in first year awards and Notices of Funding Opportunity. Changes were also noted in the topic prevalence of new awards consistent with 2025 guidelines identifying topics "not aligned with NIH priorities." These results suggest that pediatric research aimed at advancing healthcare for children is at risk with potential collateral consequences beyond childhood.

Maternal healthcare prediction systems often suffer from algorithmic biases due to socio-economic disparities and imbalanced datasets, limiting their effectiveness for equitable healthcare policymaking. This paper introduces MaternaAI, a fairness-aware and explainable learning framework designed to enhance maternal healthcare predictions in Kerala, India. The framework focuses on three critical health indicators:(1) Tetanus Toxoid (TT) booster uptake,(2) immunization coverage rates, and (3) the percentage of pregnant women completing four or more Antenatal Care (ANC) visits. To address fairness, we propose Adaptive Equity Score Optimization (AESO), a novel optimization algorithm that dynamically integrates fairness constraints into model training. AESO is model-agnostic and adapts group equity weights in response to real-time disparities. We integrate SHAP, LIME, and feature permutation techniques for explainability, enabling transparent global and local interpretation. Empirical results demonstrate that MaternaAI significantly improves fairness metrics and model accuracy across diverse machine learning and deep learning models, offering interpretable and equitable decision support for public health stakeholders.

BACKGROUND Peripheral Artery Disease (PAD) is a leading cause of lower extremity amputation. Health literacy is essential for disease awareness, but PAD awareness remains low, particularly in minoritized populations. The goal of the presented study is to investigate the PAD awareness and media preference for health information in Chinese-speaking communities in California. METHODS Anonymous 14-question surveys in Mandarin and English was designed to gauge basic knowledge of PAD, preferred methods for obtaining health information and general media preferences were collected from health fairs in San Francisco (SF), Oakland, and Los Angeles (LA) in the Chinese-speaking communities. Associations between the above variables and demographics were compared between groups. RESULTS A total of 180 responses included 94 from SF, 57 from Oakland and 29 from LA. PAD awareness was low across all cohorts. LA and SF cohorts shared similar patterns to receive health information as they preferred radio (p = 0.021, SF 49/94 and LA 8/29) while the Oakland cohort favored video/media (p = 0.024, 18/57). SF and LA cohorts showed a stronger preference for newspapers (p = 0.471, SF 32/94 and LA 12/29) and television (p = 0.244, SF 28/94 and LA 12/29), while the Oakland cohort favored video (Oakland 20/57). CONCLUSION To our knowledge, this is the first study to survey multiple US Chinese-speaking communities to assess PAD knowledge and learning preferences. The awareness of PAD among the selected Chinese-speaking communities was dismal and preferred learning methods varied from surveyed communities. Implementing community-based health education strategies will be necessary.

Type 2 diabetes (T2D) affects more than 16 million Brazilians and has nearly doubled in the last 20 years. Lifestyle interventions reduce T2D incidence in people at risk of developing the disease; however, no large-scale trials have evaluated diabetes prevention programs in Brazil or compared telehealth and hybrid delivery in middle-income settings. These are key gaps that must be addressed to enable nationwide scale-up, particularly in primary care settings and remote areas. This article presents the protocol for the PROVEN-DIA trial, designed to address these evidence gaps. PROVEN-DIA is a multicenter, open-label, randomized controlled superiority trial with a parallel design enrolling 1,305 adults with prediabetes at 30 sites across all five Brazilian regions (ClinicalTrials.gov: NCT06426277). Participants will be randomly assigned in equal numbers to one of three groups. All groups receive lifestyle guidance targeting diet, physical activity, sleep, stress, alcohol consumption, and smoking, in accordance with Brazilian national guidelines. The two intervention groups receive PROVEN-DIA, a structured 36-month lifestyle program with 43 scheduled contacts encompassing individual counseling sessions, structured support contacts, and group education sessions, delivered either in a hybrid format (PROVEN-DIA) or telehealth only (TelePROVEN-DIA). The control group receives the same lifestyle guidance through unstructured individual visits every six months, without predefined content or ongoing support. The primary outcome is the cumulative incidence of type 2 diabetes at 36 months. Secondary outcomes include body weight, fasting glucose, glycated hemoglobin (HbA1c), dietary quality, physical activity, sedentary behavior, sleep quality, perceived stress, alcohol consumption, smoking behavior, and health-related quality of life. Analyses will follow the intention-to-treat principle.

Background: Occupational therapists can improve stroke survivors hand and arm movement and participation in daily activities through action observation (AO). AO involves watching another persons hand or arm complete a movement or task. While research generally supports the use of AO with stroke survivors, there are limited AO videos are available to occupational therapists which makes applying AO challenging. Objective: The purpose of this work is to develop structured and widely accessible tool to support access to AO for stroke survivors, occupational therapists, and researchers. Methods: To develop an AO video library for stroke rehabilitation, functional and non-functional upper limb task deficits were first identified through clinical observations and clinician interviews to establish a prioritized list of daily activities. In collaboration with media production specialists, healthy adult volunteers were recruited and filmed performing these tasks from both first- and third-person perspectives. The recorded videos were then systematically edited, enhanced with instructional title slides, and distributed via a public YouTube channel for clinical application and a categorized digital repository for research purposes. Results: Initial assessments revealed a complete lack of familiarity, awareness, and utilization of AO resources among local occupational therapists, despite high perceived clinical utility. To address this gap, a final library of 150 tasks was established, resulting in the production of 419 finalized, standardized videos featuring six healthy volunteers. For clinical application, these videos were hosted on a free, public YouTube channel organized into 18 functional playlists, while a parallel set was structured into distinct movement categories for research repository storage. Conclusion: By providing a structured and highly accessible tool, this repository enables clinicians, researchers, and caregivers to readily implement evidence-based action observation interventions in both clinical and home settings.