Mathematical Biosciences and Engineering

Mosquito-borne diseases pose a growing public health challenge as climate change reshapes vector population dynamics. West Nile virus (WNV), transmitted between birds and Culex mosquitoes, disproportionately affects Maricopa County, Arizona, one of the nation's highest-burden counties, yet whether models that include weather and avian dynamics improve forecast accuracy remains unclear. Using a 15-year weekly time series of mosquito abundance, mosquito infection prevalence, and human cases, we developed four mechanistic model configurations of varying complexity, from mosquito-human dynamics alone to full models incorporating avian dynamics and weather forcing. We fitted each model to the weekly-observed data, generated probabilistic 1- and 2-week-ahead forecast horizons, and evaluated forecasts against a historical baseline. All configurations fit the data equally regardless of weather or avian dynamics. However, models incorporating both birds and weather created more accurate forecasts of mosquito abundance and mosquito infection prevalence, and all configurations outperformed the baseline for forecasting human cases. Forecast accuracy was highest in summer and fall, and ensemble aggregation sometimes outperformed every individual model, stabilizing predictions across the 15-year record. These findings indicate that avian and weather dynamics are most critical for predicting mosquito-specific data, positioning this framework as a scalable tool for public health planning for WNV surveillance under climate change.

Acute kidney injury (AKI) is a frequent complication within intensive care units, with its sudden onset often missed. This is especially important because a timely window for intervention is required as delayed detection leads to progressively worse outcomes. Existing machine learning and deep learning models have contributed to closing this gap, but their complexity, requiring hundreds to thousands of features, and lack of generalisation pose a limitation that prevents them from being integrated into clinical workflows across different electronic health-record ecosystems. This study presents a 37-feature XGBoost model trained on the MIMIC-IV dataset with 5.4% positive cases, with hyperparameters optimised via Optuna and probabilities calibrated using isotonic regression, designed for transportability across clinical settings. Validation was conducted internally using a temporal patient-level split simulating prospective deployment, training on 2008-2016 data and testing on 2017-2022 data"External validation was performed on the eICU Collaborative Research Database, a multi-centre dataset spanning 208 US hospitals, using the trained model without retraining. SHAP TreeExplainer was used to provide feature-level explainability for individual predictions. Internal testing yielded an AUROC score of 0.794 for predicting AKI onset within a 12-24 hour window. External validation produced a 0.750 AUROC without retraining. Equitable discrimination was observed across gender, age, chronic kidney disease presence, race, and AKI stages on both datasets, with a 95% internal CI of 0.789-0.799 confirming the model's estimate stability. These results suggest that clinically useful prediction systems are achievable with substantially fewer features than current models require.

Background Dengue is a major public health challenge, and predictive models are crucial for early warning systems. However, many current modeling practices rely exclusively on climatic factors or employ complex algorithms that lack the interpretability needed for informed public health decision-making. To address these shortcomings, we developed and validated a multidimensional, interpretable statistical model to predict monthly dengue incidence. Methodology/Principal Findings We used a Generalized Linear Mixed Model (GLMM) with a Negative Binomial distribution to analyze 14 years (2010-2023) of spatiotemporal data from 37 municipalities in Huila, Colombia, an endemic region. The model integrates non-linear and lagged effects of climatic, demographic, and socioeconomic factors. The final model underwent rigorous external validation on an independent test set (2021-2023). Our model demonstrated high predictive discrimination (R2 = 0.743, Spearman's {rho} = 0.657), accurately capturing the timing of epidemic outbreaks. Key findings include the identification of an optimal thermal window for transmission at 27-28{degrees}C, a threshold effect for precipitation above 800 mm, and a saturation dynamic in outbreak autocorrelation. Conclusions/Significance This mechanistically-informed statistical approach provides a robust and transparent tool for epidemiological surveillance, successfully balancing high predictive performance with the explanatory power needed for effective, data-driven public health interventions.

Drought stress is the major yield limiting factor in upland rice production where the moisture availability is highly variable. Understanding and evaluating how upland rice responds to drought stress is critical to improving resilience and yield stability. In this study performance of sixteen upland rice varieties were evaluated under non-stressed, moderately stressed and highly stressed conditions. Drought stress was introduced by irrigating upland rice at 70% and 50% field capacity (FC) whereas non-stress treatment was irrigated at 100% FC. Irrigation in moderately stressed and highly stressed conditions was also withheld for six days at lateral crop stages to observe temporary wilting by inducing a stress interval. Data on agronomic traits of upland rice was collected in three experimental replications. Results indicated that performance of upland rice varieties was significantly altered under stress conditions and highest performance was observed under non-stressed conditions. Yield losses for short duration and long duration varieties ranged 35-60% and 24-62% under moderate stress whereas it ranged 43-78% and 52-73% under highly stressed conditions, respectively. Overall varieties Dawk Kha, Khao/ Sai and Dawk Pa-yawm, indicated higher stability under stressed conditions therefore, these long duration varieties could be used for obtaining better yields under diverse agroclimatic conditions and under unpredicted weather patterns. Short duration Ma-led-nai-fai and long duration Goo Meung Lung and Bow Leb Nahag could be used for acquiring traits for higher tillering and panicle bearing capacity. Short heighted varieties such as Jao Daeng, Sahm Deuan and Ma-led-nai-fai could be used in breeding for short heighted new varieties to overcome lodging concerns. Strong significant association of GMP, STI, MPRO, MHAR with grain yield under non-stressed, moderately stressed and highly stressed conditions indicated that these indices were appropriate for their use as selection criteria for drought resilience.

Inappropriate antibiotic use presents a major global health challenge, particularly in low-resource settings where access to quality care is limited but antibiotics remain relatively unrestricted. This study estimates the causal effect of frontline primary care quality on inappropriate community antibiotic use, combining detailed community-based data from approximately 100 rural villages in rural China with an instrumental variable (IV) approach embedded within a double/debiased machine learning (DML) framework. We linked objective measures of village doctor clinical practice quality, measured through unannounced standardized patient visits, to household-level antibiotic use data collected from the same villages. To identify the causal effect, we constructed multiple candidate instruments from extensive provider characteristics and used an ensemble of machine learning algorithms within a flexible DML-IV framework to approximate an optimal instrument, addressing a many-weak-instruments problem. We found that improving village provider clinical practice quality reduced both antibiotic receipt during healthcare encounters for common diseases and household antibiotic storage for future self-medication. Our findings suggest that strengthening frontline primary care quality can meaningfully reduce inappropriate community antibiotic use without restricting access to essential treatment. More broadly, this study illustrates how causal machine learning can strengthen conventional causal estimation in complex observational settings in global health economics research.

Background: Traditional real-world studies comparing SGLT2 and DPP4 inhibitors on renal outcomes rely on propensity score matching, which causes high-dimensional data loss. We used causal machine learning (Causal ML) to unmask heterogeneous treatment effects in diabetic kidney disease (DKD). Methods: Using data from 4,588 patients within the Japanese J-CKD-DB-Ex registry, we implemented a doubly robust (DR) learning framework (Linear DR-learner with XGBoost) to compare SGLT2 and DPP4 inhibitors. Outcomes included the chronic eGFR slope and a composite renal endpoint ([≥] 50% eGFR decline or end-stage kidney disease). Heterogeneity was explored via causal SHAP and decision trees. Results: At the population level, SGLT2 inhibitors modestly slowed chronic eGFR decline (average treatment effect [ATE] = 0.14 [95% CI: -0.86, 1.15] mL/min/1.73m^2/year) and reduced composite endpoint risk by 9% (ATE: -0.09 [-0.11, -0.08]) versus DPP4 inhibitors. However, individual-level counterfactual analysis suggested that for the chronic eGFR slope, non-glinide users with stable pre-treatment trajectories who were also taking ACE inhibitors had a greater benefit from SGLT2 inhibitors (ATE: 2.95 [-0.68, 6.58]). Conversely, glinide users with steep pre-treatment decline had a greater benefit from DPP4 inhibitors (ATE: -8.98 [-16.11, -1.85]). For composite renal events, SGLT2 inhibitors had a 28% absolute risk reduction within the algorithmically identified high-risk subgroup (eGFR [≤] 28.1 mL/min/1.73 m^2 and positive proteinuria; ATE: -0.28 [-0.33, -0.23]). Even non-proteinuric decliners demonstrated a 8% risk reduction with SGLT2 inhibitors (ATE: -0.08 [-0.10, -0.06]). Conclusion: Causal ML advances precision medicine in DKD, shifting from uniform prescribing to individualized, data-driven therapy targeting distinct intrarenal pathways.

Forecasting influenza hospitalizations informs public health preparedness, yet questions remain about which types of forecasts best guide action. We evaluate categorical trend forecasts, which communicate probabilities of upcoming increases or decreases in epidemic trajectories, submitted to CDC's FluSight Forecasting Challenge between Fall-2024 and Spring-2026. Teams submitted probability distributions over five categories describing direction and magnitude of week-over-week changes in laboratory-confirmed influenza hospital admissions. We assessed performance using Ranked Probability Skill Score, Brier Skill Score, and measures of forecast-observation agreement. Most models outperformed an equal-probability baseline; the FluSight ensemble ranked among the top three in the 2024-25 and 2025-26 seasons. Forecasts were most accurate during stable periods and least during periods of rapid change, with most models underestimating observed trends. Conclusions were robust to choice of scoring metric and reference model. These results support categorical trend ensembles as an approach to communicating infectious disease forecasts that may inform public health decision-making.

Improvements in data availability, sharing, and integration, together with the development of explainable artificial intelligence (XAI) techniques, are advancing precision medicine for pediatric cancer by facilitating diagnosis, biomarker discovery, and drug development. Data sharing commons and initiatives like the Childhood Cancer Data Initiative (CCDI) provide access to pediatric-specific genomic and clinical data cohorts and improve data availability for pediatric cancer research. Based on CCDI, a scalable AI platform, Graph Artificial Intelligence for Pediatric Oncology (GAIPO), integrates various data modalities from bulk and single-cell omics data to clinical information. Such multi-modal data facilitates the training and development of advanced XAI models for pediatric cancers. We then developed an end-to-end multi-modality framework, PCGS, for pediatric cancer by incorporating omics-specific representation learning via GNN models with cross-attention fusion and multi-objective learning for downstream tasks such as classification, clustering, and survival analysis. This framework outperforms previous supervised multi-omics integration baseline approaches based on glioma and Wilms tumor cohorts and enables GNN model explainability via Shapley value-based feature attribution approaches to explain the contributions of gene-level features across various biomedical tasks, including classification and survival. Given specific background samples (e.g., age groups, sex, grades) as baselines, this explainable GNN model estimates and ranks the importance scores for input features from each omics modality. It identifies background-specific key features for biomarker discovery, risk group identification, and survival analysis in glioma and Wilms tumor, with potential applicability to other pediatric cancers.

The 17th Ebola outbreak in the Democratic Republic of the Congo, announced on 15 May 2026, was attributed to Bundibugyo ebolavirus (BDBV). Although case isolation is the main control strategy, its effectiveness is compromised when patients escape isolation facilities before recovery. Between 14 May and 17 June 2026, 175 individuals reportedly left isolation facilities without formal discharge across Ituri Province. We assessed how this "isolation leakage" affects community transmission. We refined the SEIHFR framework to distinguish undetected community infections, detected but not-yet-isolated cases, isolated individuals, leakage, funeral-associated transmission, and removals. Using Bayesian inference, we fitted the model to daily Ituri surveillance data, escapee counts, and isolation census records. We estimated the leakage rate, reporting and detection probabilities, and the transmission rate, while fixing other parameters based on the BDBV literature. The model reproduced confirmed cases, deaths, discharges, and escapees. We estimated R_0=3.67 (95% HDI: 2.0-5.7), a leakage rate of {rho} {approx} 0.034 day^-1 (0.022-0.051), and high contact-tracing-driven detection (p_d {approx} 0.91-0.99). Leakage increased the detection-dependent reproduction number [R](p_d) from approximately 3.2 to above 5. Eliminating leakage reduced cumulative infections by about one-third, from 1,120 to 764, while the minimum detection level required for control increased from p_d [≥] 0.73 without leakage to p_d [≥] 0.87 at the fitted leakage rate. Shortening time to isolation prevented the most infections (73.4%; 59-84), followed by reducing leakage (29.7%; 14-52) and re-isolating escapees (12.6%; 6-24). Delaying leakage reduction until week 4 reduced its benefit from about 27% to below 2%. Isolation leakage represents a major transmission pathway that has until now gone largely unmeasured. While rapid initiation of isolation is highly beneficial, it cannot compensate for permeable isolation; therefore, early, community-driven efforts to control leakage, embedded within a multilayered response, are critical.

Outcomes for high-grade pediatric brain tumor patients remain poor, but there is optimism that chimeric antigen receptor (CAR) T cell therapy can improve prognosis. We present the results from a phase I clinical trial of IL13BBz-CAR T cells infused weekly into the cerebral ventricles in pediatric and young adult patients with recurrent or refractory brain tumors. The trial met its primary objectives of feasibility, safety, and tolerability, with one dose-limiting toxicity. 8 of 16 patients evaluable for response experienced radiographic size decreases consistent with biologic activity and with an anti-tumor response. Two patients met protocol criteria for response. Median survival for patients receiving lymphodepletion was 20.5 months from diagnosis and 6.9 months from treatment for patients with midline glioma, and 187 months from diagnosis and 7.5 months from treatment for patients with ependymoma. Importantly, patients who did not receive lymphodepletion developed anti-CAR humoral and cellular immune responses detectable in the CSF and peripheral blood, whereas patients receiving lymphodepletion had no evidence of CSF anti-CAR immunity. Taken together, these findings demonstrate the safety, tolerability, and biological activity of locoregionally-delivered IL13BBz-CAR T cells for children and young adults with CNS tumors. Moreover, we show that anti-CAR immune responses arise in patients not receiving lymphodepletion, but not in the CSF of patients receiving systemic lymphodepletion. Further investigation of adoptive cellular therapies combined with immunosuppression is warranted in this patient population. ClinicalTrials.gov registration: NCT04510051.

Importance: Maternal vaccination and long-acting monoclonal antibodies are now available in the U.S. to prevent RSV. Long-acting monoclonal antibody administration in the U.S. commonly occurs after hospital discharge in outpatient settings, leaving some infants unprotected early in life when severe RSV risk is highest. Comparative effectiveness between the two interventions and whether delays affect effectiveness estimates have not been quantified. Objective: To evaluate the effectiveness of infant long-acting monoclonal antibody strategies and a maternal vaccination strategy, each compared to no intervention, and the comparative effectiveness of intervention strategies when accounting for real-world delays in monoclonal antibody receipt. Design: Cohort study using target trial emulation to compare four strategies for prevention of RSV-related outcomes. Setting: The U.S. between 2023 and 2025 using a nationwide database of employer-sponsored commercial insurance claims. Participants: 120,586 commercially insured mother-infants, whose infants were born in the U.S. during the 2023-2024 or 2024-2025 RSV season. Infants who could not be paired with their mother's record, did not enroll in commercial insurance within 75 days from birth, received palivizumab, and had an implausible birth date were excluded. Interventions: Comparison of four RSV prevention strategies: (i) maternal RSVpreF; (ii) long-acting monoclonal antibody given within the first week of life (mAb as intended); (iii) long-acting monoclonal antibody given within a six-month grace period from birth (mAb within grace period); and (iv) a control. Main outcomes and measures: Effectiveness against first RSV-associated hospitalization and medically-attended RSV illness was summarized using adjusted hazard ratios (aHR) and estimated using an inverse propensity weighting approach, with weights accounting for maternal age, maternal comorbidities affecting pregnancy, obstetric and newborn complications, season, region, and birth timing relative to October 1. A weighted Kaplan Meier estimator was used to estimate strategy-specific cumulative incidence of RSV outcomes over time. Results: In the first five weeks of life, the mAb within grace period strategy doubled the hazard of RSV hospitalization (aHR: 2.0 [95% CI: 1.0-4.9]) and increased the hazard of medically-attended RSV (aHR: 1.6 [95% CI: 1.0-2.7]) compared to the maternal RSVpreF strategy. The hazard for RSV hospitalization was similar for the mAb as intended strategy compared to the maternal RSVpreF strategy (aHR = 0.9 [95% CI: 0.3-1.9]). Conclusions and relevance: RSVpreF and monoclonal antibodies were similarly effective when monoclonal antibodies were administered close to birth, but when accounting for real-world delays in monoclonal antibody receipt, the maternal RSVpreF strategy was more effective than the mAb within grace period strategy.

Background Rural-urban disparities in SARS-CoV-2 infection epidemiology remain poorly quantified and understood in Zambia despite differences in healthcare access, services and preventive interventions. This study examined the geographical distribution and associated factors of SARS-CoV-2 cases across selected rural and urban districts of Zambia. Methods A convergent mixed-methods study comprised of quantitative survey and qualitative interviews was conducted in; Ndola (Urban), Kafue (Peri-urban) and Lufwanyama (Rural). The proximate determinant framework guided variable selection and interpretation. Quantitative combined (Hospital-surveillance data with community survey), while qualitative included In-depth interviews. Participants were sampled using multistage sampling technique. Quantitative data were analysed using STATA version 17, while qualitative data were analysed thematically. Findings were integrated through triangulation. Results A total of 528 participants were included, with a median age 31 years (15-71). Overall SARS-CoV-2 positivity was 12.6%, varying across rural (16.5%), peri-urban (14.9%), and urban (9.9%) settings, though residence was not associated with infection (P<0.132). Participants aged [&ge;]49 years had significantly higher odds of infection (aOR=8.78; 95% CI:1.15-66.99), whereas secondary education (aOR=0.37; 95% CI:0.16-0.86) and hospital-based testing (aOR=0.37; 95% CI:0.15-0.92) were associated with lower odds of infection. Vaccine uptake was highest in urban areas but was not independently associated with infection. Qualitative findings revealed marked rural-urban differences in perceived susceptibility, testing access, vaccine decision-making, and adherence to preventive measures, explaining several quantitative observations. Conclusion SARS-CoV-2 infection across rural and urban settings in Zambia was influenced by demographic, behavioral, and health-system factors rather than geographic residence alone. These findings highlight the need for context-specific prevention strategies, equitable access to testing, strengthened community surveillance, and targeted risk communication to improve preparedness and response for future respiratory disease outbreaks.

This study aimed to quantify the longitudinal tracking and cross-sectional construct validity of a single-item questionnaire measuring recreational physical activity frequency (RPAF) in the Womens Healthy Ageing Project. At baseline, 474 participants aged 45-55 reported RPAF from 1993 to 2014. Longitudinal tracking of the RPAF item was assessed as a consecutive-wave and baseline-referenced measure using linear weighted kappa (LWK), Spearman correlations, exact agreement and within-one-category agreement. Construct validity in the form of convergent and known-group validity was assessed using the International Physical Activity Questionnaire (IPAQ) leisure activity domains, Short Form 36 physical function (SF-36-PF) subscale, Timed Up and Go (TUG), hand grip strength (HGS) and waist-to-height ratio (WHtR). 474 participants provided baseline RPAF data. Pairwise longitudinal samples ranged from 176 to 459 across the study. Consecutive-wave LWK ranged from 0.38 to 0.49, and Spearman correlations ranged from 0.44 to 0.57. Exact and within-category agreement ranged from 41.4%-50.8% and 72.0%-79.0%. Baseline-referenced LWK ranged from 0.22 to 0.47, with Spearman correlations of 0.29 to 0.56. RPAF correlated with total IPAQ leisure score (rs = 0.60), IPAQ walking score (rs = 0.58), SF-36-PF (rs = 0.33) and TUG score (rs = -0.25). No significant correlation was identified between RPAF, HGS or WhTR. RPAF discriminated known groups for WHO guideline-sufficient activity, SF-36-PF, and TUG fall risk. The RPAF item demonstrated fair-to-moderate agreement in consecutive waves, with weaker baseline-referenced tracking. Cross-sectional validity was highest with total IPAQ leisure activity. The item may provide a pragmatic measure for RPAF in womens cohort studies.

Abstract Preventable neonatal deaths remain a major public health problem in Zimbabwe, where near-universal antenatal and facility-delivery coverage coexist with a rising neonatal mortality rate. This study evaluated whether institutionalising three core "vital signs" of the community health system (a trained village health worker (VHW) workforce, functional community governance structures, and modified women's and men's participatory learning and action groups) reduces preventable neonatal deaths in Mashonaland West Province. An embedded QUAN (qual) mixed-methods design was used, with a two-arm, parallel-group cluster-randomised controlled trial as the dominant strand. Fifty-two ward-level clusters were randomised 1:1 to the institutionalised community health system package or to standard Ministry of Health and Child Care community services, and 984 pregnant women were enrolled between 1 September 2020 and 31 October 2021, with each mother-infant pair followed to 28 days after delivery, yielding 973 mother-infant pairs for intention-to-treat analysis. The primary outcome was neonatal death within 28 days of life, expressed per 1,000 live births. The primary analysis used a three-level mixed-effects log-binomial regression model with cluster and community-health-worker random intercepts, adjusted for pre-specified covariates. Supervised machine-learning classifiers with leave-one-cluster-out cross-validation, Cox proportional-hazards regression, and multilevel logistic models were fitted as supplementary analyses. An embedded longitudinal process evaluation used key informant interviews and focus group discussions, which were analysed thematically and integrated with the quantitative findings. The neonatal mortality rate was 44.8 per 1,000 live births in the intervention arm versus 110.1 per 1,000 in the control arm. The adjusted risk ratio for neonatal death was 0.43 (95% CI 0.26-0.70; p < 0.001), a 57% relative reduction, with a number needed to treat of 16 mother-infant pairs (95% CI 11-29). Low birthweight (<2,500 g), birth interval under two years, and low community women's literacy were the strongest risk factors, while trained VHWs, functional community governance, early antenatal care, and sustained participatory group attendance were independently protective. The women's and men's groups were protective in a dose-dependent manner, becoming significant at four or more cycles (about 14 meetings) (adjusted odds ratio 0.71; 95% CI 0.60-0.85; p = 0.001). A random forest classifier discriminated against neonatal deaths with a cross-validated area under the curve of 0.842 and a sensitivity of 0.912. Qualitative findings converged with the trial results, identifying male engagement, earlier care-seeking, danger-sign literacy, social-network activation, and community death audits as the behavioural and structural mechanisms of change. Institutionalising the community health system package (trained VHWs, functional governance, early antenatal engagement, and sustained participatory groups) was associated with a substantial reduction in preventable neonatal deaths. The findings suggest that in high-coverage, high-mortality settings, the binding constraint is structural rather than clinical, and that scaling functional community governance and workforce infrastructure in the most disadvantaged communities may accelerate progress toward neonatal survival targets. The principal limitations are a one-year follow-up period, the rarity of neonatal death, and concurrent national programming that only partially reached the control clusters. Trial registration: Pan African Clinical Trials Registry, PACTR202607591142118 (https://pactr.samrc.ac.za/TrialDisplay.aspx?TrialID=PACTR202607591142118); registered retrospectively on 7 July 2026.

Biological aging occurs heterogeneously across individuals and organs. However, current measures of biological age incompletely capture organ-specific differences in health and disease risk. Because chest CT visualizes multiple thoracic organs, it offers an opportunity to quantify structural aging across organ systems. Here, we developed MOSAIC-Age, a framework characterizing eight organ-specific aging clocks on chest CT. The clocks were developed and validated using 9,971 CT scans from CT-RATE and MIDRC, and subsequently locked and applied to two independent prospective cohorts with 35,293 participants from the National Lung Screening Trial and Genetic Epidemiology of COPD study. CT-derived biological age gaps (BAGs) were examined in relation to lifestyle and socioeconomic factors, prevalent comorbidities, incident chronic diseases, and all-cause and cause-specific mortality. Higher BAGs, indicating organs that appeared older on CT than expected for their chronological age, were broadly associated with adverse health characteristics, chronic disease burden, and increased mortality risk. Multiple disease outcomes were associated with aging across several organs, whereas in multivariable analyses including all eight organ-specific BAGs, the remaining associations were more organ specific. A greater number of markedly older-appearing organs and a faster pace of aging were each associated with higher mortality. Together, these findings demonstrate that routine chest CT captures both shared and organ-specific patterns of biological aging and establish CT-derived organ aging as a quantitative imaging biomarker for assessing multi-organ health and long-term disease risk.

Stroke remains a leading cause of mortality and long-term disability worldwide, yet rapid diagnosis is often limited by the shortage of trained radiologists, particularly in resource-constrained settings. Automated analysis of CT imaging offers a potential solution, but existing methods often struggle to achieve clinically generalisable performance while jointly addressing multiple diagnostic tasks. Here we present the Intelligent Integrated Stroke Diagnosis System IISDS, an end-to-end deep learning framework built upon StrokeGNN, a graph-based architecture that integrates 3D contextual feature extraction with U-Net-based 2D lesion segmentation to enable comprehensive stroke analysis from non-contrast CT scans. IISDS performs stroke subtype classification, lesion segmentation and lesion volume estimation within a unified pipeline. To develop and validate the system, we collected and curated BGD-ISD through a collaboration between AI researchers, neurologists, radiologists and clinicians, resulting in a large multi-centre dataset comprising 1,507 CT scans from 597 stroke cases acquired across six hospitals and medical centres in Bangladesh. Across BGD-ISD and multiple publicly available datasets, IISDS achieves state-of-the-art performance on all tasks, improving segmentation accuracy by [&ge;]0.011 Dice score, reducing lesion volume estimation error by [&ge;]0.3 average symmetric surface distance (ASSD), and increasing classification performance by [&ge;]0.018 area under the receiver operating characteristic curve (AUC) compared with existing approaches. These results demonstrate the potential of graph-based deep learning to enable clinically generalisable, automated and scalable stroke diagnosis from CT imaging, supporting rapid clinical decision-making, particularly in healthcare environments with limited access to expert radiological interpretation.

Background: Aerobic exercise is recommended during stroke rehabilitation to improve cardiorespiratory fitness and support recovery; however, participation rates remain low. While institutional and system-level barriers have been widely examined, less is known about how individual patient factors influence engagement in aerobic exercise during rehabilitation. Objectives: We aimed to determine whether depressive symptoms, apathy, self-efficacy and outcome expectations for exercise, perceived barriers, or past exercise history were associated with aerobic exercise participation in stroke rehabilitation. Methods: In this prospective cohort sub-study, adults admitted to in- or out-patient stroke rehabilitation at three urban hospitals completed validated questionnaires assessing depressive symptoms, apathy, exercise self-efficacy, outcome expectations for exercise, perceived barriers to being active, and premorbid exercise history. Participants were separated into two groups for analysis: those who completed aerobic exercise during rehabilitation and those who did not. Equivalence testing and between-group comparisons were performed. Results: Sixty-two participants were enrolled; 16 participated in aerobic exercise and 46 did not. Groups were not equivalent on any individual-level factors. Compared to non-participants, those who performed aerobic exercise had significantly higher depressive symptom scores (p=0.0025) and lower self-efficacy for exercise (p=0.0087). Non-participants demonstrated significantly higher apathy (p=0.0007). No significant differences were found for outcome expectations, perceived barriers, or exercise history. Conclusion: Depressive symptoms and lower self-efficacy did not impede aerobic exercise participation during rehabilitation. Increased apathy, however, was associated with non-participation. Findings highlight the need for individually tailored aerobic exercise prescriptions that consider motivational and affective factors to optimize engagement during stroke rehabilitation.

Background and Objectives: Recurrence following burr-hole drainage of chronic subdural haematoma (cSDH) occurs in 10-25% of cases, sustained by neovascularisation of the subdural neomembrane supplied by the middle meningeal artery (MMA). MMA embolisation reduces recurrence; whether incidental burr-hole intersection of MMA branches during drainage confers similar benefit is unknown. Methods: We performed a multicentre retrospective cohort study of consecutive adults undergoing burr-hole drainage for cSDH at two UK tertiary neurosurgical centres. Postoperative thin-slice CT was used to classify burr-hole intersection of the underlying MMA groove (no hit, distal-branch hit or main-branch hit) and measure perpendicular burr-hole-to-MMA-groove distance. Co-primary outcomes were radiological recurrence and recurrence requiring intervention. Patient-clustered multivariable logistic regression adjusted for prespecified clinical covariates and treating site. Results: 227 patients (284 operated hemispheres) were included. Radiological recurrence decreased from 34.4% with no branch hit to 22.9% with main-branch intersection, with the gradient confined predominantly to unilateral cSDH. Main-branch intersection was associated with lower adjusted odds of radiological recurrence in unilateral cSDH (adjusted OR 0.30, 95% CI 0.11- 0.81; P = .018), with a similar but non-significant association in the overall cohort (adjusted OR 0.53, 95% CI 0.26-1.07; P = .075). Burr-hole-to-MMA-groove distance demonstrated a more consistent association: in the overall cohort, each 5-mm increase independently increased the odds of radiological recurrence (adjusted OR 1.38, 95% CI 1.04-1.82; P = .025). In unilateral cSDH, each 5-mm increase was independently associated with both radiological recurrence (adjusted OR 1.45, 95% CI 1.03-2.04; P = .034) and recurrence requiring intervention (adjusted OR 1.52, 95% CI 1.05-2.20; P = .027). Conclusion: Main-branch intersection of the middle meningeal artery during routine burr-hole surgery is associated with lower recurrence of unilateral cSDH, while the accompanying burr-hole-to-MMA-groove distance gradient provides biologically plausible support for a dose-response relationship. Together, these findings provide mechanistic rationale for prospective evaluation of intentional neuronavigation-guided MMA targeting (BURR-MMA; NCT07549893).

Disseminating study results to participants is an ethical responsibility for researchers but remains uncommon in low- and middle-income countries, and participants preferences for receiving study results are poorly understood. This study examined study result dissemination preferences among pregnant women in a phase III malaria prevention trial in Papua New Guinea (PNG). Participants completed an interviewer-administered questionnaire (survey) assessing their interest in and motivation for receiving trial results and preferences for dissemination methods and content. Associations between participants characteristics and dissemination preferences were explored using multivariable logistic regression analysis. Of 1172 trial participants, 96.0% (1125/1172) completed the survey, and of these 99.6% (1121/1125) wanted to learn about the trial results. The main motivation factors driving participants interest were an acknowledgment of their contribution to research (51.7%; n=579) and a better understanding of the study (45.0%; n=505). Most participants (78.9%; n=884) wanted to learn about the trial findings through written summary and a group meeting with other participants at the nearest clinic (31.1%, n=349). Multivariable regression analysis indicated that participants from rural/peri-urban clinics were more likely to choose non-electronic media dissemination approaches such as a group meeting as compared to urban-dwelling participants. Frequently selected items (>50% of participants) for content included information regarding good results of the study, purpose of the study, medical treatment advances, results specific to me, and how study was conducted. There was heterogenicity in the preference for dissemination content: compared to urban clinics rural clinics are less likely to want to learn about how and why study was conducted and medical and scientific advances. Overall, the majority wanted to learn about trial results, highlighting the importance of integrating dissemination into research activities in PNG. Variation in preferences for mode and content of dissemination between study clinics suggests that dissemination activities could be tailored to local context and preferences.

Background: Fibrosis, characterised by excessive accumulation of collagen type I (COL1), is a common feature of chronic diseases, including liver diseases (LDs), chronic kidney disease (CKD) and heart failure (HF). COL1 degradation products can be detected in urine by proteomics/ peptidomics analyses and may serve as non-invasive biomarkers of fibrosis. We aimed to identify a common molecular signature of fibrosis across these diseases that may ultimately guide interventions to slow disease progression and prevent organ damage. Methods: Using capillary electrophoresis coupled to mass spectrometry (CE-MS), naturally occurring COL1 degradation products (peptides) in the urine of patients with fibrotic disease, LDs (n=127), CKD (n=263) or HF (n=187), were investigated and compared with the same number of matched controls. Disease-associated COL1 peptides were identified separately for each condition, and peptides showing consistent associations across the three diseases were selected to define a common fibrosis signature. A support vector machine model based on the selected peptides was developed and validated in independent cohorts of patients with LDs (n=110), CKD (n=93), HF (n=32) and controls (n=643). Results: We identified a common fibrotic signature consisting of 50 COL1 degradation products, mainly downregulated in fibrosis. A model based on these peptides achieved a strong performance, with an area under the receiver operating characteristic curve (AUC) of 0.935 (95% confidence interval (CI) 0.917-0.953, p<0.0001) in an external validation cohort comprising pooled disease groups (LDs, CKD, and HF) and controls. Performance was maintained in LDs, CKD and HF, with AUCs of 0.917 (95% CI 0.890-0.944, p<0.0001), 0.951 (95% CI 0.931-0.971, p<0.0001) and 0.950 (95% CI 0.903-0.997, p<0.0001), respectively. The model scores were significantly associated with fibrosis stage in LDs (p=0.0097) and with interstitial fibrosis and tubular atrophy in CKD (p=0.045). Conclusion: A model of urinary COL1 peptides captures a shared collagen degradation signature across organs and diseases, enabling the non-invasive assessment of fibrosis irrespective of its origin. As these peptides exclusively reflect collagen degradation, the findings suggest impaired collagen degradation as a driver in fibrosis. Future clinical studies are warranted to evaluate the utility of this model for early fibrosis detection and earlier implementation of anti-fibrotic interventions.