Mathematical Biosciences

Public genomic repositories can support antimicrobial resistance (AMR) surveillance, but unequal sampling can bias interpretation. We characterised AMR determinants, multicountry genomic cluster overlap and surveillance gaps across Africa using an NCBI Pathogen Detection snapshot retrieved on 24 August 2026 for 55 African Union member states. Records were validated and deduplicated by BioSample, and complete AMRFinderPlus calls were summarised across five United Nations M49 subregions and eight overlapping regional economic communities (RECs). Country-pair cluster overlap was assessed using the Jaccard index, while project-based and composition-standardised sensitivity analyses evaluated repository bias. The dataset contained 86,829 unique BioSamples from 51 states; South Africa, Malawi and Kenya contributed 55.8%. Complete extended-spectrum {beta}-lactamase calls were detected in 21,513 isolates and carbapenemase calls in 4,642. blaCTX-M-15 dominated the ESBL profile, while NDM and OXA types predominated. Seventy clusters contained carbapenemase-positive isolates from at least two countries. A shared REC covered all participating countries in 38 clusters, while 32 crossed REC boundaries. Normalised country-pair overlap was low, with a maximum Jaccard index of 9.5%. Project balancing reduced the Northern African carbapenemase estimate from 32.3% to 17.9% and the Eastern African ESBL estimate from 36.9% to 12.5%. Public repositories identify determinants and clusters for investigation but do not estimate prevalence or transmission. AMR surveillance should combine national confirmation, regional institution-led investigation where countries share an REC, and continent-wide coordination through Africa CDC for cross-REC signals, supported by representative One Health sampling, standardised metadata and sustained African sequencing capacity.

Background Childhood respiratory infectious deaths are partitioned across four Global Burden of Disease cause modules-26 etiological attributions within lower respiratory infections, tuberculosis, COVID-19, and whooping cough-never jointly reported. Whether the structure of this combined mortality spectrum has changed over time, and with what implications for intervention design, has not been quantified. We assembled and analyzed the integrated spectrum for children and adolescents aged 0-19 years, 1990-2023. Methods We integrated Global Burden of Disease Study 2023 (release v8352) estimates into a 29-node spectrum-26 lower respiratory infection etiologies plus tuberculosis, COVID-19, and pertussis-globally and across seven super-regions, with uncertainty propagated by summing bounds. We computed Shannon diversity, Herfindahl concentration, and effective cause counts; phenotyped pandemic-window collapse and rebound per cause; linked pathogen shares to WHO/UNICEF vaccine coverage; and mapped geographic concentration in sub-Saharan Africa and South Asia. Reporting follows GATHER. Results In 2023 the 29 causes jointly accounted for 965,330 deaths (95% uncertainty interval [UI] 680,096-1,342,437). Shannon diversity rose from 2.336 to 2.711 (+16.1%) between 1990 and 2023; the effective number of causes nearly doubled (5.57 to 9.94), inversely coupled to total deaths (Spearman rho = -0.997). Whooping cough ranked second (112,954 deaths; 95% UI 64,576-185,708; 11.7%) and showed the spectrum's only rebound above 100% (-57.4% collapse, +111.0% rebound). Tuberculosis ranked third (87,764; 57,779-124,912; 9.1%) with the highest concentration in sub-Saharan Africa and South Asia (87.1%). COVID-19 entered at rank five (52,899; 47,275-59,183; 5.5%). Nineteen of 29 causes exceeded the poverty-lock threshold (>80.59% of deaths in sub-Saharan Africa plus South Asia). Conclusions Childhood respiratory infectious mortality has become more diverse and more concentrated in poverty as it has declined. Single-pathogen interventions now address a shrinking share; the spectrum's structure argues for platform interventions-oxygen, antimicrobial access, referral-tailored jointly by age and geography, implying that pathogen-specific strategies alone cannot finish the remaining mortality agenda.

Vaccine effectiveness (VE) estimates can vary widely between years and populations, even for the same vaccine. Estimated VE is known to be sensitive to susceptible depletion and differences in pre-vaccination infection risk between vaccinated and unvaccinated populations. However, how variation in pre-vaccination risk within and between the two groups affects VE estimates over time remains unclear. This uncertainty is especially important given negative VE estimates. We investigated the difference between estimated VE and true vaccine protection considering continuous distributions of pre-vaccination infection risk under three scenarios. When the vaccinated and unvaccinated populations differ in their mean risk, estimated VE can be higher or lower than true vaccine protection. Similar patterns arise when both populations share identical means but different risk distributions. Finally, if infection-derived immunity lasts longer than vaccine protection, annual VE estimates can vary by tens of percentage points between years despite constant true vaccine protection. These theoretical results underscore that VE studies estimate contrasting risk between vaccinated and unvaccinated individuals in a particular time and place, and VE estimates can vary counterintuitively between years and populations even with constant vaccine-induced protection. Explaining variability in estimated VE thus requires a more complete understanding of populations' distributions of infection risk.

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.

We propose two new tests to detect drug effects (DE) in trials of one to very few patients followed during two periods (before and after initiation of a treatment). Both methods use longitudinal natural history data to inform the estimation of each patient's DE. The first method uses a non linear mixed effect model (NLMEM) reflecting an expected natural history with a hypothetical drug effect, to estimate the Conditional Distribution of the Drug Effect (CDDE). The second method trains a Pareto Depth Analysis (PDA) algorithm, a machine learning based approach based on outlier detection, that we implement using data simulated under the NLMEM. We evaluated the two tests with a simulation study. We used data from the PROSPAX study in Autosomal Recessive Cerebellar Ataxias (ARCAs, to derive a NLMEM for the Scale for the Assessment and Rating of Ataxia score. The CDDE method provided controlled type I error and, in some scenarios, adequate corrected power, though sensitivity analyses showed vulnerability to misspecification. The PDA method demonstrated lower statistical power except with high score precision. These results highlight different strategies for quantifying treatment effects in ultra rare, patient' specific trials. They can inform methodological design for future ARCA precision therapies.

A dental AI system and a dentist reading the same radiographs form a paired comparison. Published comparative studies often report the two arms separately against a reference standard, leaving the joint pattern of correctness between them unavailable for secondary paired inference. We show what that omission costs. The accuracy difference remains exactly identified; its sampling variance does not, so the report contains the estimate and not its uncertainty. On a study of 282 units, two published accuracies are consistent with 38 distinct joint tables whose confidence intervals differ in width by a factor of 2.5. The consequence is a three-zone decision map rather than a single threshold: differences at or below 1.06 points are non-significant under every compatible table, differences at or above 6.03 points are significant under every compatible table, and in between the published numbers cannot decide. We then show the omission is repairable at negligible cost. One additional integer, the number of units both arms classify correctly, identifies the joint table exactly and restores standard paired inference. For a panel of readers the pairwise dependences must arise from one joint distribution, a constraint that binds once three readers are present; publishing each reader's joint-correct count against a single reference reader cannot widen and may tighten every pairwise bound, and in a 7-arm experiment reduced them by a median of 37% even for pairs excluding that reference. Where the integer was never published we give DentalPair-Cert, an interval with finite-sample coverage uniformly over every admissible within-unit AI-dentist dependence under the independent-sampling-unit model, certified in both the nuisance maximization and the inversion. Across 4,200,000 simulated comparisons an independence analysis falls to 74.5% coverage with 12.2% type-I error; in a purposive sample of 9 recent comparative studies, 1 reported a paired test on discordant units.

Background: Immunochemotherapy (ICT) is considered standard in regards to care for small-cell lung cancer (SCLC) in extensive stages, yet reliable biomarkers for treatment response remain elusive. While previous univariate analyses suggest specific peripheral lymphocyte subsets correlate with survival, the systemic immune response involves complex, multivariate interactions that require advanced analytical approaches. Methods: This paper analysed high-dimensional flow cytometry data from 32 patients with stage IV SCLC treated with carboplatin, etoposide, and atezolizumab. Peripheral blood was analysed at baseline (V0) and longitudinally during treatment. To identify potential early predictive biomarkers and mitigate sample attrition in later cycles, we focused on baseline and measurements after two cycles of ICT (V1). We employed a rigorous machine learning framework utilising nested cross-validation, bootstrapping, and permutation-based statistical testing to evaluate eleven different regression and survival models. Results: Under model-appropriate metrics, regressors did not generalise (R2 <0); conversely, censoring-aware Random Survival Forests (RSF) successfully extracted robust prognostic signatures. Baseline immune profiles (V0) achieved a concordance index (C-index) of 0.66 (p= 0.015), while dynamic changes from V0 to V1 ({triangleup}V) achieved a C-index of 0.65 (p= 0.022). Crucially, absolute values measured after two cycles of ICT (V1) yielded no significant signal (p= 0.445). Feature importance analysis confirmed the prognostic value of Th17 normalisation and identified Naive Regulatory T cells and Memory B cells as candidate components. Conclusion: Machine learning validation confirms a predictive signal in the peripheral immune profile of SCLC patients. Early dynamic shifts in the balance between regulatory and effector immune arms are associated with prognosis, contrasting with the lack of signal in absolute counts after two cycles of ICT. These findings establish a proof of concept for multivariate liquid biopsy immune profiling, warranting confirmation in larger cohorts and highlighting the necessity of integrating systemic and tumour-intrinsic data.

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 ([&ge;] 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 [&le;] 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.

Large language models perform well on medical examinations, but users routinely challenge their answers and invoke professional roles, and it is unclear what a system does when a medical credential and a stated task-specific accuracy point in opposite directions. In a factorial experiment on 480 items from four Polish specialty examination sets and three consumer large language model systems (ChatGPT, Claude, Gemini), each item and system received eleven independent conversations. Conditions crossed attributed source role (medical student, experienced specialist), stated prior accuracy on similar questions (2/10, 8/10) and suggestion correctness. The primary outcome was adoption of a prespecified incorrect option when the baseline answer matched the official key, comparing a specialist described as 2/10 with a student described as 8/10. Baseline agreement with the key was 87.2% across 15,683 analyzable conversations. The incorrect option was adopted more often from the specialist described as 2/10 than from the student described as 8/10 (10.2% vs. 7.6%; adjusted risk difference +2.82 percentage points, 95% CI +0.65 to +4.99). Estimates varied across the three systems and only one system-specific interval excluded zero. In a prespecified exploratory analysis with a shared eligibility rule, correct suggestions were adopted far more often than incorrect ones (risk difference +35.7 percentage points, 95% CI +30.8 to +40.7), indicating selective rather than indiscriminate compliance. An incorrect suggestion from a specialist with low stated accuracy was therefore slightly more influential than the same suggestion from a student with high stated accuracy, although the difference was modest and varied across systems. Agreement reached only after a user has disclosed a preferred answer should not automatically be treated as an independent second opinion, and medical large language model systems should be evaluated on how they revise answers after such disclosure, not solely on initial accuracy.

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 [&ge;] 0.73 without leakage to p_d [&ge;] 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.

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 widespread adoption of clinical large language models (LLMs) introduces significant risks of automation bias, premature closure, and clinician deskilling. Current interpretability paradigms, including latent space trajectories, Concept Activation Vectors, and Concept Bottleneck Models, suffer from topological stagnation, metric distortion, and epistemic occlusion, frequently masking intermediate diagnostic uncertainty behind falsely confident outputs. To address these structural vulnerabilities, this paper introduces a novel closed-loop, multi-agent framework designed to quantify and visualize dynamic epistemic uncertainty in clinical LLM reasoning. By coupling predictive Shannon entropy with non-linear Isometric Feature Mapping (ISOMAP), the architecture projects high-dimensional inference state vectors onto a calibrated two-dimensional latent space, thereby assigning a quantifiable thermodynamic energy state to the reasoning path to track diagnostic velocity, cognitive momentum, and trajectory efficiency across sequential diagnostic rounds. Pilot validation across representative emergency medicine scenarios demonstrated distinct topological and information-theoretic behaviors: unconfounded cases (cerebellar infarction) exhibited smooth geodesic progression toward the ground truth alongside monotonic Shannon entropy decay from 2.15 to 1.74; noisy environments with ambiguous findings (spontaneous pneumothorax) suffered from trajectory wandering, local minimum traps, and high sustained entropy (~2.41) due to insufficient repulsive weighting for negative evidence; and triage-conflicted cases (acute cholangitis) achieved precise geometric proximity to the true node but experienced top-1 rank stagnation because the model conflated acute severity triage (sepsis) with anatomical etiology. By rendering machine hesitation and cognitive divergence visually auditable before final diagnostic crystallization, this geometric-information framework enables dynamic trust calibration and human-AI co-regulation at the point of care while establishing a clear mathematical foundation for future architectural interventions, such as dual-channel safety decoupling and non-linear repulsive weighting. Moving forward, validating these architectural enhancements across large-scale electronic health record databases and prospective clinical trials will be essential to realize its full clinical utility, establishing a foundational blueprint for safe, transparent, and cognitively synergistic AI integration in future medical practice. By rendering the LLM's reasoning process visually auditable, this framework lays the groundwork for capturing and externalizing the clinician's own cognitive patterns within the AI, forming a coupled system. This enables the explicit visualization of cognitive gaps between physician hypotheses and AI inferences, transforming the interaction from simple answer-checking into a dynamic learning process for both human and machine that prevents diagnostic oversight. Ultimately, because the responsibility for final clinical decision-making remains with the human practitioner, this framework serves as a vital decision-support mechanism. Moving forward, validating these architectural enhancements across large-scale electronic health record databases and prospective clinical trials will be essential to realize its full clinical utility, establishing a foundational blueprint for safe, transparent, and cognitively synergistic AI integration in future medical practice.

Contact inhibition of proliferation (CIP) enables epithelial tissues to self-regulate growth and maintain tissue homeostasis. However, how cell-level mechanical contact, tissue-scale structural order, and proliferation kinetics interplay remains a fundamental open question in living matter physics. Here, we present a particle-based model of a confluent epithelial monolayer governed by overdamped dynamics, where individual cells interact via a two-dimensional hard core- soft shoulder potential. By comparing structural evolution during quasistatic densification with previously reported experimental division kinetics, we find that the dynamics of proliferation arrest mimics the onset of direct steric contacts between the hard cores of the shell. Identifying hard core contacts as the physical driver of CIP, we couple our mechanical model with a stochastic Monte Carlo division scheme in which the instantaneous division rate decreases to zero from an intrinsic value as the number of hard core contact increases to six from zero. We demonstrate that for high intrinsic division rates, the cellular densification outpaces mechanical relaxation. This kinetic mismatch drives premature hard-core contact formation, shifts the onset of jamming and contact inhibition to lower packing fractions, and induces increasingly disordered transient configurations before the tissue universally converges to a hexagonal close-packed limit. Our model's predicted division kinetics and structural order evolution are consistent with epithelial monolayer experiments, both reported and our own. This minimal physical framework links single-cell steric contact mechanics directly to tissue-scale growth regulation and structural evolution.

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.

Bayesian borrowing of external information can improve trial efficiency, particularly in pediatric and rare disease settings where patient populations are limited, but may introduce bias and inflate the Type~I error rate when the trial differs from external studies. Recent U.S. Food and Drug Administration (FDA) draft Bayesian guidance emphasizes careful evaluation of external information, prior specification, and assessment of operating characteristics. This paper compares three meta-analytic-predictive (MAP)-based methods for Bayesian borrowing: the MAP prior, robust MAP (RMAP) prior, and self-adapting mixture (SAM) prior. An adaptive platform trial design in Rett syndrome is used as a case study. Simulation studies evaluate frequentist operating characteristics under varying prior--data conflict, between-study heterogeneity, treatment effects, and clinically significant differences (CSDs) for the SAM prior. The MAP prior achieved the greatest efficiency when external and current data were compatible but exhibited the largest bias under substantial prior--data conflict. The RMAP priors improved robustness through fixed robust-component weights, whereas the SAM prior adaptively adjusted borrowing and was less sensitive to prior--data conflict while retaining efficiency gains when the data were compatible. Although the CSD influenced the degree of adaptive borrowing, as reflected by effective sample size, it had only a modest impact on frequentist operating characteristics. Sensitivity analyses using a skeptical robust component yielded similar qualitative conclusions, while accentuating the differences between the MAP and RMAP priors. These findings provide guidance for evaluating and selecting MAP-based borrowing strategies before trial implementation, particularly in rare disease settings, consistent with current FDA recommendations.

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.

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.

Human deep-space missions face bone-kidney risks that cannot be extrapolated from six-month ISS data. We built a 12-state Ca-bone-urine-stone mechanistic ODE model and jointly calibrated its 11 physiological parameters on eight ISS targets by Bayesian identification (M0 base = 19-D; M1 extension adds a GCR-bone coupling term for parsimony testing only), then propagated the M0 posterior to four environments (ISS, Lunar subsurface, Lunar surface, Mars). Lumbar-lower BMD loss increases with mission duration and partial-gravity unloading (ISS 180 d -4.83% -> Mars 730 d -12.15%; 2^3 factorial: duration 82.9%, gravity 12.5%, GCR main effect ~ 0), whereas stone rate follows the opposite gradient (ISS 16.1 vs Mars 13.1 per 1000 person-years), reflecting weakened partial-gravity bone resorption alongside residual urinary chemistry changes. The dominant pathway thus shifts from bone-centric on the ISS to kidney-centric on Mars, where residual urinary-chemistry changes-not bone resorption-drive stone risk. The direct GCR-bone coupling term is unidentifiable at current ISS doses (DeltaWAIC = +0.0076 +/- 0.126 SE), so M0 is retained as the main inference model. Bisphosphonates provide >=84% BMD protection but leave a urinary-chemistry residual, so bisphosphonate monotherapy would underestimate Mars stone risk; potassium-magnesium-citrate combinations (RRR_RSS 51%) should therefore be added to deep-space countermeasures. A Lunar-surface 365-day mission is the earliest environment on the NASA roadmap to cross a composite RED threshold. That profile differs from the regolith-shielded 180-day case in both cumulative GCR (~69x) and duration (2x), so a shielding-specific effect cannot be isolated here; forcing the GCR coupling terms to zero leaves all four composite tiers unchanged (0/4, Supp S24), and the shielded 180-day profile is YELLOW rather than GREEN. Independent hold-out validation (Culliton 2025 60-day HDT-bedrest RCT, n=8 control arm of n=24 total) supports the M0 posterior predictive distribution on the lumbar-BMD sub-scope.

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.

All primary healthcare centers owned by the Ugandan government in the Western Region of Uganda were submitted to a questionnaire concerning current availability of POCT from the Essential diagnostic List 2 part 1a and 1b, and the African laboratory inventory done by African Society of Laboratory Medicine and AfricaCDC. The data from the questionnaire was then linked to open source geodata provided by TomTom, and population data to calculate and visualize the accessibility of captured POCT. Findings: Availability of POCT Malaria is almost 100%, HIV 68-90%, and >30% for a majority of the POCT in the EDL-2 panel. 90% of the population in Western Region live within 1 hour by car from most of the essential POCT. Figures in the complementary web-based application visualize the accessibility of POCT for Western Uganda. Diagnostic deserts are visualized. Interpretation: Access to POCT at primary health care facilities in western Uganda has expanded substantially over the past decades. The geo-mapping tool presented here could inform policy decisions on strengthening diagnostic capacity at the national, regional, and provincial level. Funding: Swedish Research Council and Infravis All supplementary materials and a preprint of this submission are available on our OSF repository https://osf.io/j7puk/.