Patterns

Whole-genome sequencing comprehensively captures coding, non-coding and structural variation in families with suspected inherited disorders, yet its clinical utility remains constrained by an interpretation bottleneck: selecting a handful of relevant variants from millions of candidates. Current rule-based pipelines, anchored in ACMG/AMP criteria, excel at identifying highly penetrant Mendelian alleles but frequently miss variants of low-to-moderate penetrance, non-coding alterations and germline-somatic interactions. Here we introduce PolyCLIP-T, a topology-guided multimodal framework that transforms variant selection from a classification problem into a geometric discovery task. By contrastively aligning DNA-sequence embeddings with functional annotations, PolyCLIP-T constructs a unified latent space in which the displacement between reference and alternate embeddings quantifies the molecular perturbation induced by each variant. Persistent homology then identifies stable topological components - coherent variant groups shared among affected relatives - that transcend single-variant scoring logic. Applied to six families with multi-morbid cancer, autoimmune and cardiovascular disease, PolyCLIP-T recovered non-coding and structural candidates overlooked by conventional pipelines and revealed pleiotropic networks spanning disease categories. This approach provides an interpretable, scalable solution for genome-first investigations of disorders driven by polygenic architectures that evade single-variant analysis. The framework was developed and benchmarked on deeply characterised familial cohorts selected for transgenerational multimorbidity; validation in larger, independent populations will be essential to establish its generalisability. An interactive web tool is freely available at https://www.polyclip-t.uma.es/.

Post-surgical adverse outcomes, including mortality, intensive care readmission, and complications, remain major challenges for clinical decision-making. Existing machine learning approaches focus on outcome prediction while operating as opaque systems, limiting clinical trust and the translation of predictions into treatment decisions, and many clinical studies rely on synthetic data in which shared intermediate variables create circular dependencies between inputs and targets that compromise reported performance. We aimed to develop an explainable multimodal architecture and a rigorous evaluation methodology that address these gaps. We designed a two-stage architecture integrating supervised deep learning for risk prediction with conservative Q-learning for action recommendation. The first stage uses five modality-specific encoders for structured records, physiological time-series, chest radiographs, clinical notes, and surgical metadata, unified through cross-modal attention into a shared patient-state representation. The second stage applies offline reinforcement learning to recommend clinical actions while preventing value overestimation. We formally characterized a target-leakage flaw in synthetic pipelines and propose a real-data methodology using a verified clinical database, with event-censored temporal separation and uncertainty-weighted per-task training. Component-level behavior was validated on a controlled synthetic benchmark, demonstrating that the architecture functions as designed without claiming clinical validity. The cross-modal attention and risk-prediction components behaved as expected, whereas the offline reinforcement learning stage did not converge on the benchmark, indicating that value estimation requires further investigation on real clinical data. The architecture provides dual-level explainability through attention visualization and value decomposition, contributing a deployable design, a formal methodological critique of synthetic-data practices, and a complete framework for clinically valid evaluation.

Childhood diarrheal disease remains a leading cause of morbidity and mortality among children under five years in sub-Saharan Africa, particularly in settings affected by inadequate sanitation, climate variability, malnutrition, and limited healthcare access. Conventional forecasting approaches are often constrained by sparse surveillance data, weak spatial representation, and limited incorporation of mechanistic disease dynamics. This study presents a Physics-Informed Multimodal Artificial Intelligence Digital Twin framework that integrates Physics-Informed Neural Networks, Graph Neural Networks, diffusion-reaction epidemiological modeling, multimodal fusion learning, and Digital Twin simulation to estimate and predict childhood diarrheal disease burden in Kenya, Somaliland, and Zimbabwe. Using public epidemiological, environmental, climate, sanitation, and synthetic proof-of-concept datasets, the framework modeled temporal disease dynamics, spatial transmission, pathogen-attributed burden, and outbreak trajectories while enforcing epidemiological consistency through physics-informed optimization. Results demonstrated robust forecasting performance, enhanced spatial transmission modeling, uncertainty-aware predictions, and realistic outbreak simulations across the three countries. Rotavirus, Shigella, and Cryptosporidium were identified as major contributors to modeled mortality burden, while unsafe water exposure, poor sanitation, malnutrition, and climate-sensitive transmission substantially increased disease risk. Compared with a Bayesian baseline model, the multimodal framework achieved superior nonlinear risk characterization, geospatial learning, and temporal prediction. These findings highlight the potential of scientific machine learning and digital twin systems for infectious disease surveillance, outbreak forecasting, climate-health analytics, and evidence-based public health decision-making in low-resource African settings. Keywords: Physics-Informed Neural Networks, Graph Neural Networks, Digital Twin, Childhood Diarrheal Disease, Epidemiology, Kenya, Somaliland, Zimbabwe, Scientific Machine Learning, Spatial Epidemiology, Multimodal Fusion

Medical discrimination may alter how patients relate to health information sources following adverse care encounters. We examined whether discrimination experience is associated with selective erosion of institutional health trust and with compensatory digital health engagement, using nationally representative data from the Health Information National Trends Survey (HINTS) 6 (2022; n=6,252) and HINTS 7 (2024; n=7,278). Survey-weighted modified Poisson regression estimated prevalence ratios (PRs) for binary high-trust outcomes, and survey-weighted ordinary least squares estimated coefficients for continuous outcomes; jackknife replicate weights (50 replicates) provided variance estimates. Discrimination was associated with substantially lower probability of high trust in the healthcare system (PR=0.39; 95% CI 0.30-0.52) and physicians (PR=0.85; 95% CI 0.77-0.94), with no significant association for trust in scientists, government, family, or religious organisations. The clinical-institutional pattern replicated in HINTS 6, which additionally showed reduced trust in scientists for race/ethnicity-based discrimination. Contrary to a disengagement hypothesis, discrimination-exposed adults showed higher probability of online health information seeking (PR=1.06), health app use (PR=1.11), and online provider messaging (PR=1.13); these associations persisted after adjustment for trust in physicians. Discrimination was independently associated with lower health self-efficacy (b=-0.271). Medical discrimination selectively erodes trust in clinical institutions while leaving broader epistemic trust largely intact. Despite this, discrimination-exposed patients engage more actively with digital health channels, consistent with compensatory reorientation toward non-clinical information sources. These findings describe engaged but institutionally alienated patients, with implications for restoring clinical trust and for equity-centred digital health design.

Electronic health record (EHR) prediction models often summarize longitudinal histories as static patient-level features, which may omit potentially informative event ordering. We developed a simplified spike-timing-dependent plasticity (STDP)-inspired framework that represents asynchronous EHR data as sparse, directional transition features. The approach encodes whether one clinical event precedes another within prespecified temporal windows, preserving event identity, directionality, and approximate timing while retaining feature-level interpretability. We evaluated this framework in two retrospective prediction tasks with different temporal scales: incident acute kidney injury (AKI) prediction in 17,351 MIMIC-IV ICU stays and early postoperative recurrence prediction in 713 CUMC patients with pancreatic ductal adenocarcinoma (PDAC). Models were compared with static burden features (demographics, comorbidities, raw lab measurements) and in addition with STDP transitional feature sets using patient-level cross-validation and rolling prediction horizons. In AKI, a calibrated STDP ensemble model showed higher discrimination than static burden alone at the 24-hour decision snapshot for AKI by 72 hours, with AUROC 0.838 versus 0.800, and at 48 hours for near-term AKI prediction, with AUROC 0.868 versus 0.827. In PDAC, STDP transition features modestly improved Day -30 preoperative recurrence prediction, with AUROC 0.611 versus 0.587 and AUPRC 0.323 versus 0.318 for static burden and showed similar performance at Day 0 (7 days before recorded surgery date), with AUROC 0.681 and AUPRC 0.363. Decision-curve and feature analyses suggested that selected temporal transitions were clinically interpretable across renal, inflammatory, hepatobiliary, hematologic, glycemic, and nutritional trajectories. These findings suggest that STDP-inspired transition features may provide a practical, interpretable way to incorporate temporal ordering into EHR-based risk prediction across both acute and longitudinal settings

Since the U.S. 2013/14 influenza season, the CDC's FluSight Challenge has provided a platform for evaluating influenza forecasting models and fostering collaboration across institutions. The Challenge aims to improve the science and enhance the utility of infectious disease forecasts for public health decision making. We analyzed ten years of submitted forecasts (2014/15-2019/20 (influenza-like illness seasons) and 2021/22-2024/25 (hospital admissions seasons)) across a range of model types, including statistical, mechanistic, machine learning, and hybrid models. Influenza-like illness (ILI) forecasts were evaluated using the exponentiated logarithmic score (skill metric) while hospital admissions forecasts were evaluated using the log transformed relative Weighted Interval Score. Corresponding potential performance differences were assessed using Wilcoxon rank-sum tests, and associations with team participation history were evaluated using Spearman's rank correlation. Model performance varied by season, and no single model type consistently outperformed others. In ILI seasons, statistical models generally performed better than mechanistic and machine learning models, though consistent differences were not observed in more recent hospital admissions seasons. Ensemble forecasts showed better overall performance across seasons, and the CDC's FluSight ensemble ranked among the top-performing forecasts every year. We also found a positive correlation between forecast accuracy and the number of years a team participated in the Challenge, with statistically significant associations in four seasons. These findings highlight the benefits of ensemble approaches and sustained engagement in improving forecasting performance, while also underscoring the continued value of forecast evaluation before and following the COVID-19 pandemic. Insights from the FluSight Challenge can guide future infectious disease forecasting efforts and support more effective public health preparedness.

Background: Safe reuse of multimodal hospital data for AI development is limited by the absence of reliable, context-aware deidentification across multimodal data and longitudinal patient data. Existing approaches are largely modality-specific and can indiscriminately remove clinically important information. Methods: We developed the Multimodal Anonymizer, a modular, locally deployable multi-agent framework integrating multimodal large language models, task-specific neural networks and rule-based transformations. We evaluated 16 orchestrator model configurations on a benchmark built from publicly available data and hospital data from our institution. The benchmark dataset included data from different origins: 250 MIMIC-IV patients with synthetically injected personally identifiable information (PII) supplemented with head CT, face images, handwriting, audio, German clinical-text datasets and local data. Primary outcomes were deidentification sensitivity and preservation of clinically important content; secondary analyses examined model characteristics, reproducibility, and performance against leading market and open-source solutions. Results: The best local configuration (the orchestrator being Qwen3-VL-235B-A22B-Thinking) achieved near-complete deidentification across all datasets, with per-patient sensitivity of 98.80% (95%-CI 97.20; 100), and per-PII sensitivity of 99.82% (95%-CI 99.76; 99.88). Critical clinical preservation was 99.60% (95%-CI 98.80; 100) per-patient, and clinical preservation was 99.61% (95%-CI 99.51; 99.71) per-file. All modalities achieved at least 98.30% sensitivity (lower bound 95%-CI). On our local data, the system achieved a deidentification sensitivity of 100% per-patient and per-PII; and a critical clinical preservation of 100% per-patient as well as a clinical preservation of 99.97% (95%-CI 99.91; 100) per-file. When comparing orchestrators, the leading local models were similar to proprietary models (GPT-5.2) in deidentification sensitivity while showing higher deidentification specificity. The Multimodal Anonymizer outperformed previous tools on most modalities. Conclusion: Near-complete, utility-preserving deidentification of multimodal clinical data is achievable with a unified, locally deployable multi-agent system, enabling safer large-scale reuse of hospital data for research and AI development.

Background: Machine learning models trained on population health surveys offer scalable tools for cardiovascular screening, but recurring methodological weaknesses undermine their credibility and equity: data leakage from synthetic oversampling, qualitative rather than quantitative explainability evaluation, and the absence of demographic fairness auditing at the clinical operating threshold. Methods: We present EXHEART, a leakage-free stacked ensemble pipeline trained on BRFSS 2015 (n = 253,680) and validated on BRFSS 2020 (n = 319,795; temporal transport and retrain) and a clinical cardiovascular examination dataset (n = 68,730). The pipeline combines XGBoost, LightGBM, Random Forest, and a multi-layer perceptron as base learners with 5-fold out-of-fold logistic regression stacking and Platt scaling calibration. A quantitative SHAP-LIME consistency framework, based on Kendall-tau rank correlation and Jaccard overlap, accompanies a decision-curve analysis, a subgroup-stratified SHAP interaction analysis, and an intersectional fairness audit (Sex x Age x Income) with threshold-shifting mitigation and a frontier of the fairness-utility trade-off. The framework also adds cross-instrument fairness-disparity attribution, an empirical diagnostic that provides evidence on whether an observed subgroup disparity is more consistent with a measurement-induced or a substantive explanation by re-validating it on a dataset that measures the same clinical construct objectively. On heart disease, this diagnostic associates 89% of the sex TPR gap (95% CI [0.65, 0.99]) with the self-reported survey outcome rather than with a substantive risk difference. Results: On BRFSS 2015, EXHEART achieves AUC-ROC = 0.850, AUPRC = 0.371, Brier score = 0.071, and reduces ECE by 96% (0.256 to 0.011) via Platt scaling. Global SHAP-LIME rank agreement is moderate-to-strong (Kendall-tau = 0.580, Spearman-rho = 0.818) with a substantial top-3 divergence (Jaccard@3 = 0.200), where Stroke flips from SHAP rank 8 to LIME rank 1. The Sex TPR gap is 0.124 at the screening threshold; intersectional Sex x Age disparities reach 0.649 among adequately-powered cells, 5.2x the single-attribute gap. Temporal transport to BRFSS 2020 collapses sensitivity from 0.776 to 0.267, while retraining restores AUC = 0.840 and ECE = 0.012. On clinical examination data, the Sex TPR gap collapses to 0.014; the attribution test indicates this gap is instrument-dependent, consistent with a measurement or outcome-definition explanation rather than a substantive risk difference. Cross-domain SHAP analysis identifies four instrument-independent CVD risk factors and two major portability failures. Conclusions: EXHEART combines three practices that population-scale cardiovascular classifiers usually apply in isolation: leakage-free training with calibrated probabilities, a test of whether the model's explanations are stable, and a fairness audit that examines intersecting subgroups rather than single attributes. Bringing them together proved worthwhile. The intersectional audit revealed disparities that single-attribute auditing missed, and the cross-instrument comparison indicated that much of the sex gap reflects how the outcome is measured in survey data rather than a substantive difference in risk. The temporal transport findings indicate that deployed BRFSS models warrant periodic monitoring and retraining to maintain clinical utility. EXHEART is a retrospective methodological evaluation on public de-identified data; it is not validated for direct clinical decision-making, diagnosis, or treatment recommendation without prospective clinical validation.

Large language models embedded in autonomous agents process trusted instructions and untrusted data in one context window, leaving them open to direct and indirect prompt injection. In healthcare this is not hypothetical: a 2025 JAMA Network Open study found commercial medical LLMs followed injected instructions in 94.4% of simulated patient encounters, including life threatening recommendations . Yet the clinically decisive problem we quantify here is different. Most real clinical threats protected health information PHI exfiltration, cross patient access, bulk export, out of scope advice are fluent, legitimate looking requests that carry no attack signal, so even a state of the art injection detector passes them. Existing runtime guardrails trade safety against latency: model based auditors are accurate but add hundreds of milliseconds of Python inference, while lexical filters are fast but blind to obfuscated or semantically disguised payloads. We present QFIRE, an inline, provider agnostic prompt firewall implemented as a single self contained Rust toolchain proxy, CLI, and benchmark harness. QFIRE combines three mechanisms: (i) positive security scope constraints, which restrict a model call to a declared natural language purpose and block out of scope drift even when no overt attack token is present; (ii) an asynchronous detector graph that runs N rules and their detector nodes concurrently, cheapest checks first; and (iii) a de obfuscation pass that decodes Base64 hex ROT13, folds homoglyphs and leetspeak, and strips zero width characters before detection. QFIRE ships 106 versioned firewall rules and a dedicated HIPAA Safe Harbor 18 identifier PHI panel, and runs a local DeBERTa v3 injection classifier via embedded ONNX Runtime. On 1968 public prompt injection and jailbreak prompts QFIREs deterministic hybrid attains F1 0.86, statistically tied with Metas state of the art PromptGuard 2 0.86 and above protectai DeBERTa v3 0.83; lexical baselines lag 0.16 to 0.50. Our central result is on QFIRE HealthBench, a new 2000 prompt healthcare benchmark we build and release with real garak and Microsoft PyRIT payloads. There the same PromptGuard-2 recovers only 0.40 recall DeBERTa v3 0.57, because most clinical threats carry no injection signal; QFIREs combined scope plus PHI chain reaches 0.83 recall F1 0.87 at a calibrated 0.08 false positive rate. Generic injection detection, even state of the art, is therefore necessary but not sufficient for healthcare agents. A bare LLM judge also closes most of this static corpus gap F1 0.90; QFIREs contribution beyond static accuracy is auditable determinism, bounded latency, and adaptive robustness, where the bare judge falls to 34 to 59% recall section 5.5. End to end, placing QFIRE in front of a tool using agent over a mock EHR sandbox cuts the agents harmful action rate from 0.38 to 0.00 at a 0.13 benign utility cost. All code, rules, corpora snapshots, and scripts are released, and every table regenerates from a single make paper target against local models with no paid API keys.

Ecological momentary assessment (EMA) enables real-time, longitudinal measurement of symptoms and behavior via smartphones, yet nearly all feasibility evidence comes from protocols lasting one to two weeks, far shorter than the timescales over which chronic diseases fluctuate and clinical decisions unfold. Whether daily compliance can be sustained over months, or whether it decays as short-protocol trends predict, is unknown. Here, 214 participants (173 with pain, 41 healthy controls) completed a 4-month (122-day) EMA protocol via the Soma smartphone app, generating 26,907 check-ins. Half the sample completed the full protocol without a two-week lapse. Aggregate compliance appeared moderate (50%), but this conflated two distinct phenomena: when recomputed over each participant's active period, compliance rose to 71%, with 91% achieving moderate-to-high adherence, and remained stable across all 17 study weeks. Pain status predicted earlier disengagement but not lower compliance among those who remained; after adjustment for differential retention, group differences disappeared. To our knowledge, this is the longest continuous daily EMA evaluation in a clinical population. It suggests the primary barrier to long-duration EMA is not declining motivation among active participants but concentrated early disengagement, with direct implications for the design of digital health protocols, decentralized trials, and remote symptom monitoring.

High-frequency physiological monitoring in ICUs can identify impending deterioration hours before clinical recognition yet extracting reliable early-warning signals from noisy vital-sign streams remains challenging. We present SIgnose, an interpretable prediction framework for early detection of abnormal shock index (SI), built from routinely monitored vital signs using physiologic variability and nonlinear time-series features. SIgnose was developed on the eICU Collaborative Research Database and externally validated on the MIMIC-III adult database and a pediatric SafeICU cohort (AIIMS New Delhi), with additional prospective validation in the pediatric ICU. We benchmarked three representation strategies: (i) engineered physiologic variability and nonlinear time-series features, (ii) deep learning, and (iii) Llama-3.1-8B embeddings with low-rank adaptation. Physiologic variability features consistently demonstrated superior cross-cohort generalization. The final model used 3,970 features from five vital signs to predict abnormal SI up to 8 hours ahead, achieving AUROC 0.861 (95% CI 0.859-0.863) and AUPRC 0.927 (95% CI 0.925-0.929) on eICU. External validation yielded AUROC 0.870 (95% CI 0.863-0.876) and AUPRC 0.935 (95% CI 0.930-0.940) on MIMIC-III, and AUROC 0.875 (95% CI 0.863-0.888) and AUPRC 0.915 (95% CI 0.898-0.930) on SafeICU; prospective pediatric validation (n = 88) achieved AUROC 0.885 (95% CI 0.868-0.902) and AUPRC 0.911 (95% CI 0.882-0.936). SHAP interpretability analysis identified heart rate variability, respiratory trend dynamics, and multi-scale blood pressure variability as key early-warning signatures. These findings establish SIgnose as a reproducible, low-compute, early-warning framework and demonstrate that physiologic variability features provide robust, generalizable representations for early deterioration detection across adult and pediatric critical care.

Background. Fairness-aware machine learning increasingly targets demographic performance disparities in clinical prediction, yet whether standard bias mitigation strategies genuinely improve equity in physiological signal analysis remains unclear. Age-based disparities in photoplethysmography (PPG)-based heart rate prediction present a particular challenge, as age-related performance differences may reflect context-dependent physiological structure rather than correctable artifacts. Methods. We evaluated three fairness interventions, inverse-frequency weighting (IF), Group Distributionally Robust Optimization (GroupDRO), and adversarial debiasing (ADV), applied via fine-tuning of a PPG foundation model across three clinical datasets spanning intensive care unit, laboratory, and consumer wearable contexts. Outcomes were assessed using a 2x2 framework classifying each intervention-dataset combination by the joint direction of change in mean absolute error (MAE) and fairness gap (FG) across age groups, yielding four outcome types: genuine improvement (G), leveling down (L), selective benefit (S), and both worse (W). Results. Across nine intra-domain conditions, no intervention simultaneously improved both MAE and FG (0/9 genuine improvement). The dominant pattern was leveling down (5/9): FG decreased but was accompanied by MAE degradation, indicating that apparent fairness gains were achieved at the cost of overall predictive performance. Age-group difficulty ordering varied across clinical contexts at baseline and was not preserved under intervention. In 18 cross-domain transfer conditions, genuine improvement was rare (4/18) and observed exclusively in non-MIMIC source configurations; models fine-tuned on MIMIC-sourced data yielded no genuine improvements (0/6). Embedding-level representation changes following fine-tuning did not reliably predict fairness outcomes. Conclusions. Age-based fairness interventions in PPG heart rate prediction indicate a leveling-down pattern rather than genuine equity improvement, suggesting that age-related performance gaps reflect context-dependent physiological structure not fully addressable through standard bias mitigation. Cross-domain transfer further amplifies this instability. These findings suggest that fairness evaluation frameworks for age-stratified physiological prediction should account for context-dependent performance structure rather than treating observed gaps as correctable bias.

Background. Plasma and serum metabolomic studies of myalgic encephalomyelitis / chronic fatigue syndrome (ME/CFS) have repeatedly implicated hypometabolic, lipid, mitochondrial, redox and tryptophan-kynurenine pathways, but prior cohorts have been modest in size and have used heterogeneous case definitions. Whether similar pathway-level signals are detectable at scale in dried blood spots (DBS), across questionnaire-derived fatigue constructs and across orthogonal LC gradients in the same individuals remains unresolved. Methods. We profiled DBS extracts from 1,784 community-cohort adults by reverse-phase LC-MS using paired 5 min and 15 min gradients. Six questionnaire-derived endpoints captured a pragmatic self-reported PEM-like phenotype, a DSQ-derived PEM-like construct, high or review clinical status, temporal fatigue state, comorbid fatigue and self-reported chronic fatigue. The locked primary endpoint for Phase 1 was pragmatic_fatigue_pem with 226 cases and 914 controls after excluding major metabolic comorbidity. We tested a biology-first panel comprising 22 literature-curated metabolites represented by four participant-level descriptors each, and evaluated three discovery extensions: a targeted m/z search of additional literature candidates, a hypothesis-free univariate screen across 4,553 5 min and 5,625 15 min consensus features, and pairwise z-difference ratios. Endpoint-specific Ridge classifiers were evaluated by five-fold out-of-fold AUC with bootstrap stability filtering. Cross-gradient agreement was assessed by per-metabolite AUC concordance between paired 5 min and 15 min profiles. Severity was modelled as an ordinal grade derived from the number of fatigue criteria met and chronic-fatigue-form status. Results. The biology-first DBS panel achieved out-of-fold AUC 0.81 for the pragmatic self-reported PEM-like endpoint (226 cases / 914 controls). The DSQ-derived PEM-like construct reached AUC 0.60 (57 cases / 201 controls) on the un-filtered set and AUC 0.778 (SD 0.013, twenty seeds) in a post-hoc signature-decomposition follow-up restricted to participants without a self-declared major-metabolic-history tag (29 cases / 230 controls); both are treated as construct-validity anchors rather than as provoked or clinically adjudicated PEM. An optimised operationalisation of the same construct (panel-self normalisation, restriction to non-comorbid participants and demographic covariates) reached AUC 0.71 (95 % CI 0.55 to 0.76), and an exploratory age-stratified signature decomposition suggested age-dependent pathway composition that requires confirmation given small per-stratum case counts. Stable contributors mapped to carnitine-shuttle, TCA-cycle, redox-thiol and tryptophan-kynurenine pathways. Cross-gradient analysis of 22 matched metabolites yielded Pearson r = 0.62 for signed univariate effects (p = 0.002; 68 % directional agreement). The metabolomic score increased with severity grade (Spearman rho = 0.45, p = 4 x 10^-91; median scores 0.24, 0.51 and 0.75 across grades 0, 1 and 2). Sensitivity analyses on the covariate-complete subset (n = 565; 138 cases / 427 controls) showed that the DBS signal was robust to adjustment for age, sex, BMI and medication burden (DBS-only AUC 0.76, DBS plus covariates 0.78, covariates only 0.64), and produced a metabolomic-specific lift of approximately 0.13 AUC over the strongest anti-leak declarative cross-form questionnaire baseline (AUC 0.63). DBS-only AUC was stable across sex, age and BMI subgroups, and a 1:4 nearest-neighbour matched analysis on age, sex and BMI yielded AUC 0.72 (95 % CI 0.67 to 0.77). The observed pattern supported pathway-level convergence with prior ME/CFS metabolomics literature, including carnitine shuttle, fatty-acid beta-oxidation, TCA cycle, redox-thiol, urea cycle, glycerophospholipid and tryptophan-kynurenine axes. In contrast, the hypothesis-free 15 min screen produced high-AUC features that mapped predominantly to environmental or technical signals, including pesticide, industrial-amine and mobile-phase artifact annotations; only one of eight top leads, a truncated oxidised phospholipid, was biologically plausible, and none had tandem-MS support. Conclusions. In this large community cohort, a literature-curated DBS metabolomic panel captured pathway-level biology associated with a questionnaire-derived PEM-like fatigue phenotype, showed directional concordance across LC gradients, scaled with symptom severity and remained robust to key demographic, anthropometric and anti-leak questionnaire baselines. The findings converge with several metabolic axes previously reported in ME/CFS plasma and serum studies, including carnitine-shuttle, TCA-cycle, redox-thiol, urea-cycle, glycerophospholipid and tryptophan-kynurenine pathways. They should not be interpreted as clinical validation of a diagnostic test, screening tool or objective provoked-PEM biomarker. Rather, they support at-home-compatible DBS metabolomics as a biologically grounded platform for future clinically adjudicated validation, decision-support development and longitudinal monitoring in fatigue and PEM-like syndromes. Because DBS contains cellular and plasma-derived components, matrix effects must be considered when comparing individual metabolites with venous plasma or serum studies, and hypothesis-free screening at this scale can preferentially surface exposome or technical variance unless molecular identification is enforced before biological interpretation.

Machine learning (ML) has emerged as a transformative technology across biomedical and life science sectors, with applications spanning drug discovery, medical imaging, genomics, and clinical decision support (Goecks et al., 2020; Patel et al., 2020). Despite exponential growth in ML-related publications, from fewer than 100 articles in 2003 to nearly 25,000 by 2021 (NCBI, 2022), adoption among industry professionals remains uneven and sector-dependent. Understanding what drives or inhibits this adoption is critical for organisations seeking to leverage ML capabilities in research and clinical practice. Technology adoption in organisational contexts has been extensively studied through the Technology Acceptance Model (TAM), originally proposed by Davis (1989) and subsequently extended to incorporate external variables influencing perceived usefulness (PU) and perceived ease of use (PEU) (Venkatesh & Davis, 1996). While TAM has been applied across multiple industries, its application within biomedical and life science contexts remains limited, and the industry-specific factors that shape ML acceptance in this sector have not been systematically examined. Two external variables are particularly relevant to life science professionals. First, the bibliometric journal impact factor (JIF) functions as a cognitive signal of scientific credibility, a sector where evidence-based decision-making is culturally embedded, and publication quality serves as a proxy for technological legitimacy (Garfield, 1996). Second, technology hype, operationalised through the Gartner Hype Cycle framework, represents a social influence variable that shapes organisational expectations and investment decisions around emerging technologies (Gartner Inc., 2018). Whether these variables influence ML acceptance among life science professionals, alongside individual knowledge and experience, has not been empirically tested. This study addresses that gap by investigating ML technology acceptance among 213 biomedical and life science professionals across EMEA, LATAM, and North America, using a cross-sectional quantitative survey and PLS-SEM analysis. The TAM model is extended with three external variables, JIF, technology hype, and prior knowledge and experience, to test their influence on PU and PEU in this specific professional context. Additionally, the study examines demographic and regional differences in ML acceptance, with particular attention to variation between academic researchers and healthcare professionals. The findings contribute a validated, sector-specific extension of TAM for life sciences, provide actionable insights for organisations seeking to accelerate ML implementation, and establish a framework for future subsector-specific research.

Background: Machine learning models for stroke mortality prediction typically treat each time horizon independently and use flat tabular features that ignore the relational structure of electronic health records (EHRs). In this pilot study, we leveraged graph-based machine learning models to predict post stroke all-cause-mortality across three different time horizons. Methods: We developed Stroke Temporal Heterogeneous Graph (StrokeTHG), a heterogeneous graph neural network model for simultaneous multi-horizon stroke mortality prediction (30-day, 90-day, 1-year) using EHR data from Penn State Health System. The model encodes various relations among EHR entities (e.g., patient, diagnosis, comorbidity) and temporal encoding of admission time to better predict stroke mortality. We compared our proposed approach against various baseline methods, including Logistic Regression, Random Forest, and XGBoost. We also performed ablation and subgroup analyses, evaluated the quality of learned graph embeddings, and assessed the importance of different edge types in the graph. Results: We included 4,144 stroke patients (mean age 69.2 years; 54.3% men), of whom 3,332 (80.4%) survived their stroke after one year. 30-day, 90-day, and 1-year mortality rates were 9.7%, 13.7%, and 19.6%, respectively. Our proposed approach, StrokeTHG, achieved AUROC of 0.872, 0.878, and 0.837 across horizons, outperforming all tabular baselines. At [≥] , 75% specificity, the model identified 5-10 percentage points more mortality cases than the best baseline at each horizon. Subgroup analysis demonstrated consistent performance across sex subgroups and the largest discriminative gains in the Age 65-80 stratum. Edge-type ablation identified phenotype-patient and admission-patient edges in the constructed EHR graph as the most influential relational edges for mortality prediction. StrokeTHG embeddings outperformed all graph and matrix factorization baselines under an identical downstream classifier, confirming that performance gains stem from representation quality rather than classifier capacity. Conclusions: StrokeTHG demonstrates that heterogeneous graph representations of EHR data provide a consistent improvement over flat tabular models for multi-horizon stroke mortality prediction, with particular advantage at clinically actionable sensitivity thresholds and novel multi-horizon monotonic prediction capability. This methodological framework may be adaptable to other EHR-based clinical research studies seeking to leverage heterogeneous relational structures for predictive modeling.

Background: Molecular medicine has made genetic diagnostics crucial for rare diseases, but the majority of patients remains without diagnosis even after state-of-the-art assessment. Standardized systems for integrating clinical features, such as the Human Phenotype Ontology (HPO), offer assistance, but are often insufficiently detailed and fail to capture crucial clinical parameters such as age at onset, longitudinal changes in symptoms, detailed characteristics of a clinical symptom, or the absence of a feature. Results: We present Genosolver an integrated workflow that utilizes machine learning to address this bottleneck. Using Large Language Models (LLMs) and Large Reasoning Models (LRMs) on unstructured clinical notes and electronic health care data, we generate a workflow that unifies phenotype extraction, generates differential diagnosis, and prioritizes genetic variants from genome data. We evaluated the performance on 233 previously genetically solved cases, where Genosolver ranked the causative gene first in 72% of cases and in 94% of cases in the top 10 gene list, outperforming the existing benchmarking tool Exomiser by 9%. Semi-automated reanalysis of 1,875 unsolved rare disease cases yielded an additional diagnostic rate of 1.7%. Incorporating rich, unstandardized clinical narratives substantially enhanced model performance beyond HPO-only inputs and demonstrated competitive results using data security compliant local models. Conclusion: Integrating unstandardized clinical data with local LLMs and reasoning offers a scalable, data-secure workflow that increases molecular diagnoses in rare diseases.

Background. Blood-based biomarkers are increasingly proposed for identifying high-risk individuals before clinical disease and for making prevention-oriented trials more efficient. Prognostic enrichment can increase event rates, but trial efficiency also depends on whether the intervention effect is preserved in the enriched population. Methods. Using the UK Biobank Pharma Proteomics Project, we trained disease-specific proteomic risk scores (ProRS) from 2,916 plasma proteins with elastic-net Cox models. We compared ProRS, polygenic risk scores (PRS), and combined PRS--ProRS scores across ten incident diseases. We estimated cumulative incidence and theoretical two-arm time-to-event trial sample sizes across risk strata. To evaluate effect preservation, we examined six intervention-analogue exposure--outcome pairs spanning genetic (PCSK9/coronary artery disease, APOE/Alzheimer's disease, PPARG/type 2 diabetes, IL23R/Crohn's disease), behavioural (physical activity/all-cause mortality), and pharmacological (RAAS inhibitors versus calcium channel blockers/coronary artery disease) examples. Results. ProRS outperformed PRS for 9 of 10 diseases (median C-index 0.75 versus 0.61). ProRS and PRS were weakly correlated (median Pearson |r| = 0.04), and joint PRS--ProRS stratification identified groups with higher observed incidence than either score alone for several endpoints. In the top risk quartile, combined-score enrichment reduced theoretical required sample sizes by 32--74\% under a fixed 20\% relative hazard reduction. These gains were not always preserved when stratum-specific intervention-analogue effects were used. Effects were broadly preserved for APOE/Alzheimer's disease and physical activity/mortality. The PPARG/type 2 diabetes effect attenuated toward the null under all three score types, showing that event-rate enrichment does not guarantee effect preservation. For IL23R/Crohn's disease and the antihypertensive comparison, point estimates differed across score types -- preserved under polygenic but attenuated under proteomic enrichment -- but confidence intervals were wide and overlapping. Conclusions. Proteomic risk scores can identify high-event-rate populations for prevention-oriented trials, but event-rate enrichment alone is insufficient for trial design. Biomarker-guided enrichment should evaluate mechanism-specific effect preservation and may be preferable as a stratification or adaptive-design variable rather than as a restrictive eligibility criterion.

Between 2024 and 2025, Colombia universalized the Electronic Health Invoice with embedded Individual Health Services Delivery Records (RIPS -- Registro Background Between 2024 and 2025, Colombia universalized the Electronic Health Invoice with embedded RIPS records (FEV-RIPS) as the standard for financial and clinical data exchange. ADRES -- the entity responsible for administering the resources of Colombia's General Social Security Health System -- faced the challenge of processing information from multiple heterogeneous sources generated by more than 55,000 healthcare providers. Health systems in high-income countries converge clinical-financial data in consolidated platforms; Colombia started from a fragmented architecture with incompatible historical sources, no cross-database standardization, and no centralized analytical infrastructure until 2023. Objective We describe the design, technical challenges of integrating heterogeneous data, and operational performance of the analytical infrastructure built by ADRES to centralize large-scale processing of Colombian health system information, and derive transferable lessons for health system resource administrators in Latin America facing equivalent digitalization mandates. Methods Technical-descriptive report based on operational metrics from the ADRES Azure/Databricks environment during January-November 2025. We report indicators of data volume, processing speed, computational capacity, concurrent use by functional group, and governance structure. The architecture integrates VPN connectivity with MinSalud, automated processing of multiple formats (XML, relational tables, flat files), and a medallion data lake (Bronze/Silver/Gold). Data quality challenges include structural inconsistencies across sources, coding incompatibilities (municipalities, dates, diagnoses), format heterogeneities in unstructured data, and absent technical documentation. Results The platform manages 21 catalogs, 1,183 tables, and over 110,645 million stored records, with cumulative production exceeding 1 trillion processed records. It executes queries on 100 billion records in ten seconds using clusters of up to 32 TB RAM and 4,096 vCPU. During September-October 2025, monthly query peaks reached 78,028 across eleven functional groups. Integration required Python/PySpark parsers for variable-depth XML, equivalence tables for incompatible municipality codes, cleaning routines for extreme dates used as nulls (1900-01-01, 9999-12-31), and transformation logic bridging classic RIPS and FEV-RIPS. The platform supported econometric analyses, judicial mandate responses, and public interactive dashboards. Conversational AI integration (Genie, Copilot) extends analytical access to users without SQL knowledge. Conclusions ADRES built in one year an analytical infrastructure that provides, to our knowledge, the first published documentation of the systemic technical challenges of integrating heterogeneous data sources in a middle-income social security health system. Centralizing health system information at national scale is technically feasible under public institutional constraints -- but requires solving cross-source standardization problems the implementation literature does not document with quantitative precision. The derived lessons are transferable to health system resource administrators in Latin America facing equivalent challenges.

Skin cancer requires early detection for improved survival rates. Most existing methods rely on deep learning based image classification, which is affected by visual similarity among lesions. Fewer studies use Gene Expression (GE) analysis, which captures molecular characteristics but lacks structural and visual details. To overcome limitations of individual modalities, this paper proposes a multimodal framework integrating dermoscopic images and GE profiles for skin cancer classification. EfficientNet and logistic regression are used for image based analysis and genomic skin lesion profiling, respectively, followed by fuzzy rule based decision systems to reduce uncertainty within individual modalities. Finally, fuzzy fusion combines predictions from both modalities using uncertainty based weighting of classifier outputs. The experimental findings show that both the image based and GE based classification models individually achieved accuracies of nearly 92%. However, the integration of prediction results through the proposed fuzzy fusion strategy further enhanced the classification performance, achieving an overall accuracy of 94.25%. The results obtained outperform contemporary methods, highlighting the effectiveness of combining complementary multimodal information compared with single modality approaches.

Background: Unstructured data represent about 80% of total electronic health records (EHR) data. Structuring this free text is essential for advancing clinical research, including cohort selection for trials, retrospective studies, and the development of disease registries. While manual chart review (MCR) remains the gold standard for extracting this clinical data, the process is inherently slow, resource-intensive, and susceptible to errors from human fatigue. We evaluated the extraction accuracy, safety, and efficiency of the HeLIX (Hepatology Logic-Integrated Extraction) framework, a Large Language Model (LLM) protocol using Google Gemini 3 Pro, compared to a gold-standard Manual Chart Review (MCR). Methods: A prospective validation study was conducted using 50 high-complexity, simulated hepatology discharge summaries designed to replicate the real-world heterogeneity of EHRs. The HeLIX framework employed a Zero-Shot, Structured Chain-of-Thought (CoT) prompting strategy enforced by a three-layer architecture: Clinical Reasoning Trace, Schema Enforcement, and Evidence Verification. The model extracted 45 distinct clinical variables. Performance was benchmarked against a consensus MCR. Results: Across 2,250 evaluated data points, the model achieved an overall Extraction Accuracy of 99.24% (95% CI: 98.8%-99.5%), with perfect concordance in 35/45 (77.8%) variables. For binary diagnostic variables, the model demonstrated an overall F1-score of 0.98, Recall of 0.99 and substantial inter-rater reliability (Cohens {kappa} = 0.97). Hallucinations were exceptionally rare (2/2250; 0.08%). Critical errors affecting clinical management occurred in only 2 instances (<0.1% of total data), both involving etiological misattribution in complex multifactorial diagnoses. The AI workflow was 13.4-fold faster and 95.1% more cost-effective than manual extraction. Conclusion: The HeLIX framework demonstrates physician-level accuracy and reliability in extracting complex hepatology data. It offers a scalable, efficient, and economical alternative to manual chart review. Such frameworks could accelerate clinical research, enabling healthcare systems globally to build comprehensive patient registries for a fraction of the traditional cost.