Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences

Background: In Japan, health planning is organized around secondary medical areas (SMAs; niji-iryo-ken; 330 areas in the 2025 classification), yet nationwide analyses of intensive care unit (ICU) capacity have been conducted mainly at the prefecture level, and a recent SMA-level study addressed only the presence or absence of ICUs. The full supply structure of intensive and intermediate critical care - ICU and high care unit (HCU) beds - has not been characterized at the SMA level with respect to its composition, road-network accessibility, and evolution over time. Methods: We developed MeshScope-Region, an analytical platform built on the Hospital Bed Function Reports (byosho-kino-hokoku) for fiscal years 2016-2024, in which ICU and HCU beds were identified from notified reimbursement categories and aggregated to SMAs. Three analytical layers were integrated: (1) cross-sectional distribution of ICU/HCU beds; (2) nationwide road-network accessibility computed with the Open Source Routing Machine (OSRM) from 176,962 populated 1-km census grid cells to all facilities reporting ICU or HCU beds; and (3) a nine-year longitudinal analysis of supply-structure types, classified by k-means (k = 6) in an 8-dimensional PCA space anchored to fiscal year 2024, with earlier years projected into the same space. Results: In fiscal year 2024, 20,631 ICU/HCU beds were reported nationally (7,114 ICU-type; 13,517 HCU-type) at 1,044 facilities. Zone-level totals among SMAs with any beds ranged 229-fold (3-688 beds); the 90th/10th percentile ratio of per-capita density was 3.6. In total, 90.1% of the population resided within 30 minutes' drive of a facility with ICU beds and 97.8% within 60 minutes; only 0.8% resided beyond 90 minutes. Although 140 of the 330 SMAs had no ICU facility within their own boundaries, 84.7% of their residents could reach an ICU facility in an adjacent area within 60 minutes' drive. Longitudinally, supply structures were highly persistent: 63.0% of SMAs (208/330) retained the same structural type across all nine years, adjacent-year rank correlations of a supply-vulnerability index were 0.887-0.924 (2016 vs. 2024: rho = 0.711), and the number of SMAs with zero ICU beds remained frozen at 133-141. The Gini coefficient of bed distribution declined from 0.384 to 0.262 - although computed on ICU-type beds alone it remained 0.365 in fiscal year 2024 - and capacity growth (total +27.9%) was driven predominantly by HCU beds (+41.6%) while ICU beds grew only +8.0%. Conclusions: Japan's critical care supply structure is regionally rigid, with a stable set of approximately 140 SMAs lacking ICU beds for nearly a decade, yet road-network accessibility substantially mitigates the consequences of zone-level absence. Recent capacity growth - and much of the apparent equalization - has occurred predominantly in intermediate care. MeshScope-Region provides a standing, reproducible evidence base at the geographic unit of Japan's medical planning cycles.

Background: Female chronic pelvic pain disorders (CPPDs) are highly prevalent and frequently accompanied by sleep disturbance and autonomic nervous system (ANS) dysregulation. Heart rate variability (HRV), a non-invasive index of ANS function, may provide an objective, physiological correlate of sleep health and can be monitored using wearable devices, enabling a continuous, scalable approach. Objectives: This study examined whether wearable-derived daily HRV metrics are associated with self-reported sleep disturbance in women with CPPD(s) compared with healthy controls, using epoch-level data and generalized additive models. Methods: We conducted a retrospective observational study using up to 90 days of data from a mobile health research app. Participants were 128 women with CPPD(s) and 63 demographically matched healthy controls, who completed a daily PROMIS-based 3-item sleep disturbance questionnaire and wore Fitbit devices that provided 5-minute HRV epochs. Primary predictors were high frequency (HF) and low frequency (LF) power and root mean square of successive differences (RMSSD), with group (CPPD vs control), daily pain severity, and menstrual status as covariates. We fit separate generalized additive mixed models (GAMMs) for each HRV metric with a nonlinear smooth term and an HRV x Group interaction. Results: Higher HF and RMSSD were associated with lower sleep disturbance scores, and these associations were stronger in controls than in the CPPD group (HF x group B {approx} -1.59, p < 0.00010; RMSSD x group B {approx} -0.58, p < 0.0001). LF showed a more complex pattern but also differed by group (B {approx} -0.531, p < 0.0001). HRV smooth terms were highly nonlinear, and models explained ~8-9% of deviance in sleep disturbances. Pain severity and menstrual bleeding were strongly associated with worse sleep. Conclusion: These findings indicate small but consistent associations between wearable-derived HRV metrics and daily sleep disturbances in women with CPPD(s) and healthy controls, with weaker associations in CPPD(s). Integrating continuous HRV with symptom tracking could support low-burden and multimodal monitoring of sleep health in chronic pelvic pain, but prospective validation is needed before HRV can be used for diagnostic or treatment response decision making.

We present FootNet, a 453-image multi-view smartphone foot dataset for binary foot segmentation, with expertannotated masks across six anatomical views (dorsal, medial, and plantar, both left and right). We benchmark four segmentation models under a controlled protocol: U-Net with a MobileNetV2 encoder achieves the best performance (IoU 0.9268, Dice 0.9608, 95 % CI [0.9209, 0.9320]); DeepLabV3 with MobileNetV3-Large scores IoU 0.8984 (Dice 0.9449); UNet++ with MobileNetV2 scores IoU 0.8913 (Dice 0.9391); and SAM ViT-B with oracle boundingbox prompt scores IoU 0.9219 on the matched 191-image subset. Bonferroni-corrected Wilcoxon signed-rank tests (k = 6 comparisons) show U-Net significantly outperforms DeepLab (p < 0.001, r = 0.638) and SAM ViT-B with oracle boundingbox (p = 0.005, r = 0.202); UNet++ does not significantly differ from DeepLab (p = 0.062). Connected-component postprocessing yields negligible benefit (mean {triangleup}IoU = +0.0003, 12 of 453 images improved). The extended dataset is available upon request

Abstract Introduction: Proximal humerus fractures account for approximately 5-6% of all adult fractures and are primarily managed nonoperatively. Healing is conventionally monitored with radiographs, with radiopaque callus formation indicating healing. Visible radiographic callus appears weeks after biological union begins. Ultrasound provides a dynamic, radiation-free, and cost-effective method that can detect early callus formation before x-ray visibility. Although ultrasound has demonstrated utility for fracture healing in the clavicle and humeral shaft, its role in proximal humerus fractures remains unclear. Methods: This single-centre prospective study will be conducted in two phases. The pilot phase will measure inter-rater reliability for ultrasound detection of early callus formation at 2 and 4 weeks post-injury. Ten patients with proximal humerus fractures treated nonoperatively will undergo standardized anterior and lateral scans. Each patient will generate four saved images (short- and long-axis views), producing forty anonymized images independently reviewed by two raters. The prospective cohort phase will recruit approximately thirty additional patients. Results: Reliability will be quantified using Cohens kappa. A power calculation will be performed after pilot analysis. Results from the prospective cohort phase will help determine the association and predictive value of early ultrasound-detected bridging callus for radiographic and clinical union at three and six months. Patient reported outcome measures will be assessed using the Quick Disabilities of Arm, Shoulder and Hand (QuickDASH) questionnaire. Discussion: This study will develop and validate a standardized ultrasound protocol for assessing early fracture healing in proximal humerus fractures. By establishing both inter-rater reliability and predictive value, the findings may support ultrasound as a reproducible, radiation-free adjunct to conventional imaging and enable earlier identification of union status.

Fracture nonunion remains a major cause of morbidity, yet patient-specific factors associated with impaired healing remain incompletely characterized. Anemia has been associated with adverse orthopaedic outcomes, but its relationship with fracture nonunion is poorly understood. We examined whether pre-fracture anemia, anemia burden, and clinically relevant anemia subtypes were associated with nonunion following tibial or femoral fractures. Using commercial and Medicare fee-for-service claims from 2016 through 2023, we identified adults aged 19 years or older with a tibial or femoral fracture, continuous enrollment during the preceding year and for at least six months after fracture, and no baseline cancer. Pre-fracture anemia was evaluated as any anemia, the number of distinct anemia diagnoses, and nutritional, hemolytic, aplastic, and other anemia subgroups. Nonunion occurring six to eighteen months after fracture was assessed using incidence rates and multivariable-adjusted hazard models. Among 326,673 adults, 149,704 had pre-fracture anemia and 176,969 did not. The crude incidence of nonunion was 42% higher among individuals with anemia than among those without anemia (incidence rate ratio, 1.42; 95% confidence interval, 1.32 to 1.53) and increased with greater anemia burden. After adjustment for demographic and clinical characteristics, including prior fractures at other anatomical sites, pre-fracture anemia remained associated with nonunion following tibial and femoral fractures, with hazard ratios of 1.83 (95% confidence interval, 1.54 to 2.18) and 1.38 (95% confidence interval, 1.26 to 1.50), respectively. Associations were also observed for nutritional and other anemias, whereas estimates for hemolytic and aplastic anemias were limited by few nonunion events. Within the femur, the association was strongest for distal fractures. These findings demonstrate that pre-fracture anemia is independently associated with nonunion. The increase in risk with greater anemia burden and findings across evaluable subgroups suggest that pre-fracture anemia may help identify patients at increased risk of impaired fracture healing.

Objectives To identify stakeholder-perceived design tensions in a two-sided marketplace for reusable digital therapeutics (DTx) software components and to use these tensions to propose alternative marketplace concepts. Methods We conducted 24 semi-structured interviews with digital health researchers and professionals. Data were analysed using hybrid deductive-inductive codebook thematic analysis. The Magic Triangle provided the initial deductive structure. One researcher coded all transcripts; a second independently applied the developing codebook to five transcripts to refine definitions and consistency. Seventeen parent themes were synthesized into 12 design tensions, which informed three author-generated marketplace concepts. Results Participants described trade-offs concerning target users and host, component scope and customization, quality labels, verification, geographic scope, pricing, interoperability, platform launch, risks and market niche. The resulting concepts emphasized a regional startup ecosystem, a research-oriented hybrid marketplace or a global marketplace with stricter entry requirements. Discussion The concepts combine the tensions in different ways and highlight competing priorities in governance, openness, assurance, scalability and early platform growth. Conclusion Stakeholders identified recurring design choices for a DTx software-component marketplace. The concepts provide hypotheses for prototyping and evaluation; the study did not test technical feasibility, market demand, regulatory acceptability or effects on development cost or time.

In this paper, we propose a new model to estimate the impact of an intervention on human hosts of a vector-borne infection, such as dengue, which occurs in yearly outbreaks of different magnitudes. The model applies to these outbreaks and, in fact, is independent of their intensity, that is, it does not require the steady-state assumption. The model takes as input the officially reported age-dependent number of cases of a vector-borne infection. It is deterministic and does not account for stochasticity. Our objective is to estimate the impact of the intervention (the efficacy), and we rely on the observed fact that the age distribution of the proportion of cases of the infections transmitted by the same vector is independent of both the intensity of transmission and the geographic area studied, at least for Brazilian regions. This finding is highlighted in the main text and forms the basis of our calculations. A hypothetical intervention is simulated using a dengue vaccine, which allows the determination of the optimal strategy for a vaccination campaign.

We develop a mathematical model of Rift Valley Fever integrating mosquito vectors, ruminants, and humans, based on an SEIR-type structure with vertical transmission in vectors. Local data from the Sudanian and especially the Sahelian zones are used to capture the impact of climatic variations on mosquito population dynamics. The mathematical analysis establishes the models positivity, determines the basic reproduction number R0, and demonstrates the local and global stability of the disease-free equilibrium. Sensitivity analysis (PRCC) highlights the most influential parameters, while the stochastic approach using a continuous-time Markov chain confirms the major role of seasonal rainfall. Numerical simulations reveal a peak in animal and human infections around the 9th month, correlating with periods of heavy rainfall. This model provides a relevant tool for surveillance and prevention within a "One Health" approach in Chad.

This study presents ReCo (Research Cosmos), a self-configuring and self-extending agentic research framework for the biomedical domain. ReCo is orchestrated by a large language model that interacts with native computing tools, bundled Model Context Protocol (MCP) servers, structured skills, persistent project memory, and a desktop interface. Its bundled MCP servers provide biomedical analysis capabilities while serving as implementation paradigms for integrating new computational and AI frameworks. Structured skills encode procedures for environment configuration and framework ingestion, enabling ReCo to inspect repositories, manuscripts, or local codebases; identify dependencies and execution patterns; create isolated runtime environments; design and implement MCP interfaces. Self-extension was evaluated using five heterogeneous systems: the Merlin computed tomography foundation model, MAISI-v2 medical image synthesis framework, asari liquid chromatography-mass spectrometry workflow, DosimeTron agentic radiation-dosimetry platform, and Orthanc DICOM server. ReCo successfully operationalized all five systems and completed predefined functional evaluations. Re-hosted DosimeTron outputs demonstrated near-perfect agreement with the reference pipeline across 651 organ observations (Pearson correlation and Lin concordance correlation coefficient, 0.99999; mean absolute percentage difference, 0.37%). Notably, ReCo configured Orthanc as a PACS-like coordination layer, integrated it with DosimeTron, Merlin, and TotalSegmentator, and orchestrated data retrieval, analysis, and return of valid DICOM RTSTRUCT, RTDOSE, and Structured Report. ReCo provides a unified environment for configuring, documenting, and operationalizing heterogeneous biomedical frameworks, reducing technical barriers to the adoption and integration of emerging computational and AI methods. The official open-source ReCo GitHub repository is available at: https://github.com/eltzanis/ReCo

Patients and caregivers seek informational and emotional support throughout medical care, especially when interpreting unfamiliar laboratory test results. Although resources such as patient portals and online health communities (OHCs) help address questions, gaps remain. The emergence of large language models (LLMs) offers the potential to be a complementary source of support to assist patients and caregivers in understanding and using their test results. The objective of our study is to empirically compare LLM responses to patients online questions containing their laboratory test results to responses written by peers in an OHC. We compared the 519 peer replies to 122 laboratory test-related posts from an OHC to 488 responses generated from four LLMs using mixed computational and qualitative methods. LLMs frequently provided clear explanations of medical terminology and structured interpretations of numeric results but were longer and less readable. Peers offered more personalized, context-specific emotional support. Overall, LLMs have the potential to complement peer responses in OHCs, but require greater emotional depth, reasoning transparency, and alignment with community norms.

Introduction Adolescent idiopathic scoliosis (AIS) requires frequent x-rays for management, exposing young patients to cumulative radiation risks. While radiation-sparing imaging modalities exist, access across the National Health Service (NHS) remains uneven and information given to patients is variable. This qualitative study investigated the systemic, geographic, and interpersonal dynamics of AIS imaging in England. Design This qualitative study employed in-depth semi-structured interviews with healthcare professionals (HCPs) from NHS paediatric spinal centres, patients aged 13 to 25 years old with AIS and parents/carers of young people with AIS. Setting England. Participants A total of 22 HCPs from 13/24 NHS paediatric spinal centres in England, 19 10-25 years with AIS and 11 parents/carers. Results Conventional x-ray remains the main imaging modality. Significant geographic inequality exists. The most commonly available radiation-sparing imaging modality available is the EOS system, which uses slot-scanning technology, is available at 7 centres in England, primarily in London imaging networks. Acquisition of EOS systems is currently driven by local charitable funding rather than a centralised strategy, with high capital and installation costs cited as primary barriers. Inconsistent knowledge of imaging within primary care and a lack of specialist expertise in local secondary care services led to diagnostic redundancy, gatekeeping, and low value inconsistent imaging. These systemic delays frequently closed the window for conservative treatments like bracing. A professional balancing act exists between the duty to inform and the desire to minimise patient anxiety. HCPs often use selective communication regarding radiation risks. Conversely, families demonstrate high relational trust with HCPs and low baseline knowledge of cumulative exposure, often viewing frequent imaging as a reassuring marker of clinical progress. In centres with EOS systems, clinicians felt empowered to lead proactive, transparent risk discussions. In standard X-ray settings, dialogue remains reactive and infrequent, leading to a reliance on implied rather than truly informed consent. Conclusions AIS imaging in England is variable. Geographic location dictates access to low-dose radiation technology and the quality of informed consent. Systemic inefficiencies and fragmented referral pathways contribute to diagnostic redundancy and delayed specialist care. National standardisation of clinical pathways, information provision and a centralised strategy for low-dose technology procurement are essential to eliminate structural inequalities and ensure equitable, transparent care for all patients.

Objective: Photobiomodulation therapy has emerged as a promising modality to facilitate scar healing and pain management in dermatology and plastic surgery. However, its role in postoperative care following spine surgeries remains understudied. This double-blinded, placebo-controlled study aimed to investigate the effects of photobiomodulation in patients with chronic lower back pain undergoing lumbar decompression, with postoperative wound healing as the primary outcome and pain reduction and functional recovery as secondary outcomes. Methods: Patients were randomized to receive either active photobiomodulation braces (N=13) or placebo braces (N=12). Follow-up assessments were performed at 2, 4, 6, 8, and 12 weeks postoperatively. Outcomes included wound healing (Stony Brook Scar Evaluation Scale), back and leg pain (Visual Analog Scale), quality of life (EuroQol 5D), and functional status (Oswestry Disability Index). Results: Compared to the placebo group, the photobiomodulation treatment group had a 4.12-fold cumulative improvement in final scar scores, with significant between-group differences at postoperative weeks 6, 8, and 12 (p = 0.0062, 0.010, 0.042). Among patients with severe preoperative disability, treatment resulted in a 1.89-fold faster improvement in back pain (p=0.025) and a 1.80-fold faster improvement in ODI scores (p=0.025); and superior treatment effect on wound healing were again observed at weeks 6, 8, and 12. Among patients with poor initial scars, treatment led to a significantly better scar outcome than placebo at week 6 and a 1.94-fold faster EQ5D improvement (p=0.052), with significant gains observed as early as two weeks after surgery. There were no adverse events associated with photobiomodulation treatment. Conclusions: Photobiomodulation significantly promoted postoperative wound healing following lumbar decompression surgery, with therapeutic benefits preserved even in patients with poor baseline scar scores and functional impairment. This indicates that the efficacy of photobiomodulation is not limited by the initial scar condition or disability, supporting its broad clinical applicability. Additionally, patients with severe preoperative disability experienced greater benefits from photobiomodulation than placebo, including faster reduction in back pain and more rapid improvement in functional capacity, highlighting its role in postoperative pain management and rehabilitation. These therapeutic effects are likely mediated by photobiomodulation-induced reduction of inflammation and enhancement of tissue repair. Together, this study suggests that photobiomodulation can be a promising adjunct therapy to facilitate postoperative recovery in patients undergoing spine surgery.

Objectives: To compare the effectiveness of povidone-iodine ear wash compared to no ear wash and oral cotrimoxazole compared to placebo given in addition to standard topical antibiotic treatment (ciprofloxacin drops) for chronic suppurative otitis media (CSOM) in Australian Aboriginal children. Methods: A randomised, parallel, 2 x 2 factorial design, assessor-blinded clinical trial in the remote Northern Territory of Australia. Aboriginal children with confirmed CSOM were eligible to be randomised into four treatment groups, allowing two primary treatment comparisons in a 2-in-1 trial approach. Participants received standard treatment (twice daily cleaning and topical ciprofloxacin drops) plus: i) either 16 weeks of pre-treatment povidone-iodine ear wash or no povidone-iodine ear wash; and ii) either 16 weeks of oral cotrimoxazole or placebo. Central randomisation with allocation concealment and triple-blinding of the oral antibiotic treatment arms was used. The relative risk (RR) and risk difference (RD) were estimated after adjustment for age, community, and the other intervention. The primary outcome was the proportion of children with any otorrhoea (clinical failure) after 16 weeks of treatment. Secondary outcomes included size of tympanic membrane (TM) perforation and amount of discharge, time to cessation of discharge, proportion of children with respiratory and other pathogens in ear discharge (at baseline and 16 weeks) and hearing levels (at 12 months). Findings: 280 children with CSOM were randomised and 270 had their primary outcome assessed. Clinical failure (presence of any ear discharge) after 16 weeks of treatment was 66/134 (49%) in the povidone-iodine group versus 69/136 (51%) in the no povidone-iodine group (RD= -1% (-12,11), p= 0.93) and 56/134 (42%) in the cotrimoxazole group versus 79/136 (58%) in the placebo group (RD=-16% (-28,-4), p=0.007). The amount of discharge, TM perforation size, the level of hearing impairment, and serious adverse events were not significantly different in both treatment comparisons. Anaerobic growth (24%), Pseudomonas aeruginosa (21%) and Haemophilus influenzae (17%) were the most common pathogens found in the ear discharge before treatment. Fungi or yeast (24%), Staphylococcus aureus (15%), and anaerobic growth (10%) were the common pathogens after 16 weeks of treatment, with no significant differences between groups. At 12 months post-randomisation, 55-60% of children had at least one discharging ear and there was no difference between treatment groups. Interpretation: Povidone-iodine ear washes did not contribute to better ear outcomes in this study. Cotrimoxazole for 16 weeks resulted in more children with clinical improvement to dry ears. Oral cotrimoxazole may play a role in reducing the burden of CSOM in populations with high rates of persistent disease.

Electronic health records (EHR) are increasingly recognised as critical digital infrastructure for integrated, patient-centred care in the context of rising multimorbidity. In low-resource settings, national EHRs may also support locally driven learning to improve adaptive care across chronic conditions. However, there is limited empirical evidence on whether and how these systems enable learning within routine care in ways that inform broader system adaptation. We conducted a qualitative multi-method assessment of Impilo, Zimbabwe's national EHR, to examine its capacity to support learning for integrated multimorbidity care at primary care level, using HIV-hypertension as a tracer condition pair. Guided by Friedman's socio-technical infrastructure model as the analytical framework and Learning Health Systems (LHS) theory as the interpretive framework, data were drawn from documentary review, ethnographic observation, patient journey mapping, and interviews with frontline health workers and key stakeholders. Frontline learning for person-centred multimorbidity care was actively generated through interpretation of patient trajectories, experiential adjustment, and coordination across HIV and hypertension services using both the EHR and paper-based artefacts such as registers and patient booklets. However, this learning remained largely encounter-bound and weakly stabilised. Impilo did not routinely provide usable longitudinal patient views, practice-facing analytic tools, or institutionalised mechanisms for collective reflection required to support integrated multimorbidity care. Consequently, learning was largely confined to incremental adjustment within existing workflows, with limited capacity to inform broader changes to care pathways, routines, or system design. These findings suggest that the principal barrier to developing LHS is not the absence of data or frontline learning capacity, but the lack of socio-technical arrangements that enable learning to stabilise and inform system adaptation. Digitalisation alone is insufficient to support adaptive multimorbidity care. Co-production with frontline health workers may provide a pathway for aligning digital system design with routine care realities.

Automated interpretation of neuroimaging studies requires simultaneous assessment of multiple imaging evidence variables, each tied to distinct anatomical structures. Vision-language models (VLMs) offer a unified framework for multi-task analysis, but adapting pre-trained VLMs remains challenging. Full fine-tuning is computationally prohibitive, and joint multi-task training requires simultaneous access to all task data, which is often infeasible in clinical settings. Although model merging enables multi-task composition without joint re-training, existing methods focus on post-hoc algorithms with limited extension to VLMs and minimal application to neuroimaging. Here, we present GRadient-guided Adapter Merging (GRAM), a layer-selective low-rank adaptation (LoRA)-based fine-tuning and merging framework for multi-task neuroimaging visual question-answering (VQA). GRAM uses a gradient ratio that contrasts class-specific gradients to identify task-discriminative layers, and applies subspace-constrained projected gradient descent to restrict LoRA updates to directions consistent with the geometry of the pre-trained model. We leveraged a structured VQA benchmark, developed from the National Alzheimer's Coordinating Center (NACC) dataset, that pairs multi-sequence brain MRI studies with question-answer pairs across clinically relevant imaging evidence variables. Experiments on the VQA benchmark showed that GRAM outperformed or matched all-layer LoRA fine-tuning and a standard merging baseline while reducing inter-task interference during merging, and approached or surpassed the performance of joint multi-task training without joint re-training.

Magnetic Resonance Imaging (MRI) derived brain age varies substantially between individuals, but it remains unclear whether early deviations from normal brain ageing precede future cognitive decline and whether they provide predictive value beyond conventional MRI measures. Here, we investigated whether MRI-derived brain age gap estimation (BrainAGE) identifies early structural brain ageing differences among cognitively normal individuals who later develop mild cognitive impairment (MCI) or dementia. We analysed longitudinal structural MRI data from the Alzheimer's Disease Neuroimaging Initiative (ADNI) and replicated the main findings in the population-based Kuopio Osteoporosis Risk Factor and Prevention Study (OSTPRE). Individuals who later converted to MCI or dementia had higher BrainAGE values several years before diagnosis and, in ADNI, showed steeper longitudinal increases than stable individuals. Elevated BrainAGE values were also associated with increased risk of future conversion to MCI in cognitively healthy individuals and faster subsequent memory decline. Cross-sectional differences and the association between BrainAGE and risk of future conversion were replicated in OSTPRE. Importantly, adding BrainAGE to models including demographic, APOE4, cognitive, and MRI-derived measures consistently improved prediction of future cognitive outcomes, with the greatest benefit observed for individuals who converted after longer follow-up. These findings show that structural brain ageing begins to diverge years before the onset of MCI. BrainAGE captures this early divergence, providing complementary information beyond conventional structural MRI measures that may improve the early identification of cognitively normal individuals at increased risk of future cognitive decline when integrated with other biomarkers.

Background Primary care in China lacks structured mental-health assessment, and the machine-learning models that could support such screening are typically developed on heavily selected samples. Cumulative inclusion and exclusion criteria, though usually treated as neutral data-cleaning steps, can create heterogeneity in predictive reliability among retained participants. Using the China Health and Retirement Longitudinal Study (CHARLS) 2011 baseline, we quantified how selection funnels distort epidemiological associations and inflate machine-learning metrics, and tested selective prediction as mitigation. Methods Using the CHARLS 2011 baseline with temporal external validation in CHARLS-2018, we built a four-level selection funnel (L0-L3), evaluated five classifiers with nested cross-validation and SMOTE, and compared model-embedded uncertainty with a decoupled predictor-selector framework; XGBoost cross-validation residuals drove risk stratification and classification and regression tree (CART) rules. Results Sample sizes fell from L0 n=17,705 to L3 n=4,256 (24.0%). The cancer-depression odds ratio attenuated from 1.78 (95% CI 1.32-2.41) to 1.39 (0.74-2.63), losing significance. AUC rose with selection but not after multiple-comparison correction, whereas calibration error increased for four of five models. Model-embedded uncertainty succeeded only for XGBoost; with the decoupled XGBoost residual selector, all five models achieved selective prediction at approximately 20% coverage (test AUC 0.90, 95% CI 0.85-0.95), abstaining on approximately 80% of cases for individual safety. Risk stratification was stable (residual Spearman correlations >0.95; multi-seed Jaccard 0.88), and CART rules used self-rated health, education, pain, and marital status. Conclusions The findings support a deployable primary-care triage pathway: a four-variable rule identifies patients suitable for algorithm-assisted scoring (approximately 20% coverage) and routes the remainder to human evaluation. Methodologically, cumulative selection bias produces a dual distortion: epidemiological associations are compressed and machine-learning metrics inflated. Selective prediction is limited mainly by uncertainty-indicator design. Performance metrics should be reported with selection level, coverage, and calibration trajectory. Decoupled selective prediction with CART rule extraction provides an actionable framework for quality-controlled, tiered-care deployment. Keywords: selective prediction, selection bias, CHARLS, depression, predictor-selector decoupling, uncertainty quantification, classification and regression tree, triage, clinical decision support, health management.

Background: In Japan, acute inpatient care is divided into approximately 335 secondary medical care areas, which serve as the basic units for planning healthcare delivery systems under the 8th National Health Care Plan. While comparisons between regions and facilities typically rely on a single risk-adjusted metric, this approach confuses differences in patient demographics with differences in the actual infrastructure of intensive care units (ICUs). This paper presents a framework - MedZone Embedder - for deriving data-driven indicators of regional structural vulnerability by mapping secondary medical care areas onto a learned similarity space, together with its working implementation. The paper sets out the concept, the method, a proof of concept, and an explicit staged validation program, rather than national empirical results. Methods: Each area is represented by a feature vector consisting of aggregated values of intensive care provision indicators derived directly from the Japan Intensive Care Patient Database (JIPAD) - specifically, risk-adjusted mortality rates (standardized mortality ratios and an in-hospital composite indicator), technical efficiency, length of stay, readmission rates, case severity, and case composition - with the within-area variance of these indicators also taken into account. No hierarchical processing by facility type is performed. A contrastive autoencoder (multilayer perceptron encoder 32 -> 16 -> 8, symmetric decoder) is trained by self-supervised learning, using an objective function that combines reconstruction and normalized temperature cross-entropy (NT-Xent) on noise-augmented views. The resulting 8-dimensional embedding supports area searches based on cosine similarity and anomaly scoring in the embedding space (using isolation forest, Mahalanobis distance, or k-nearest-neighbor density), which is normalized to a vulnerability score ranging from 0 to 1. If deep learning libraries are unavailable, or if the number of areas is small, an alternative method using deterministic principal component analysis is employed. Results: This method was implemented and deployed within an operational ICU decision support system on a managed cloud platform. The proof of concept (PoC) is structured around five secondary medical care areas within Kyoto Prefecture and runs entirely on synthetic facility-level aggregate data constructed to follow the JIPAD indicator schema; no registry data were accessed. It generated: an aggregate provision profile for each area; an area embedding space equipped with a similar-area search function; and a vulnerability ranking that identifies areas with low patient numbers and low diversity that exhibit overall poor outcomes. At this scale, the contrastive autoencoder falls back to principal component projection. The deep learning pathway has been implemented and unit testing has been completed; training and evaluation on actual registry data are pending data-use approval and the expansion of data integration. Validation is staged: Stage 2 will train the contrastive pathway over JIPAD-covered areas to assess construct validity against public structural indicators (ICU/HCU beds, population, accessibility), and Stage 3 will extend coverage to all areas via National Database (NDB) linkage. Conclusion: MedZone Embedder reframes regional comparison from single-indicator ranking to structural representation: which areas are alike, and which are structural outliers. The contribution of this paper is the framework - the proposal that the intensive care provision structure of Japanese secondary medical care areas can be learned from a national outcomes registry and read through the lens of what we call institutional debt - together with a deployed implementation and a pre-specified validation program. To our knowledge, this is a candidate first application of contrastive representation learning to Japanese secondary medical care areas.

Background: International Classification of Diseases (ICD) codes are often used in epidemiological studies to track disease rates over time. Objective: This evaluation of ICD-code-based algorithms for electronic medical record (EMR) studies of Alzheimers disease (AD) and related dementias (ADRD) examines the impact of incorporating Centers for Medicare and Medicaid (CMS) data as an additional source of diagnostic and treatment information in Department of Veterans Affairs (VA) EMR studies. Methods: We performed a chart review of 100 VA Million Veteran Program (MVP) participants to evaluate algorithm performance. We also assessed genetic associations across algorithms in a large MVP cohort (n=396k). Results: Adding CMS data increased the number of detected cases, sensitivity, and positive predictive value, but decreased specificity and negative predictive value. Genetic analyses showed that broader (ADRD/dementia) algorithms with just VA data performed similarly to narrow (AD-focused) algorithms incorporating both VA and CMS ICD codes. Additionally, narrow AD algorithms based solely on VA data yielded the highest ORs, indicating the largest proportion of late-onset AD cases. Conclusions: We recommend using a broad (ADRD) algorithm without CMS or medication data, particularly for epidemiological studies or a strict AD algorithm including CMS and medication cases for genetic discovery of late-onset AD associations in VA EMR, and a strict AD algorithm without CMS data for applications focused solely on AD and sensitive to misspecification. Careful evaluation of algorithm performance is warranted in different EMR systems, as ICD coding practices vary by institution, as demonstrated by this comparison of VA EMR and CMS data.

Background. Translating a clinical guideline decision algorithm into a computational graph requires judgment, and unconstrained coding yields divergent graphs; any complexity measure computed from such a graph inherits that variation, so its reproducibility must be demonstrated rather than assumed. Objective. To develop, and prospectively test, an empirical method for making graph extraction reproducible, using the Clinical Guideline Complexity Index (CGCI) and four guideline algorithms as a case study. Methods. We built a Graph Representation Specification (an ontology, a motif catalogue, disambiguation conventions, decomposition rules, a deterministic validator, and a scoring engine) and refined it by error-driven grammar induction: measure inter-coder disagreement, localize its dominant class, induce a single grammar rule, and prospectively test whether that rule improves agreement in the anticipated class. Reproducibility was quantified with a pre-specified, topology-based endpoint (Decision Topology Agreement) rather than edge agreement, which is oversensitive to representational choices that do not affect the score. Two trained coders independently coded the diabetes, dyslipidemia, heart-failure, and hypertension algorithms. Results. A rule induced from the diabetes comorbidity panel (assessment topology) generated a pre-specified prediction that heart-failure figures, sharing the same motif, would converge; on a fresh, independently coded pair they did, with an absolute CGCI difference of approximately one. Decision topology reproduced closely (decision-order agreement at or near 1.00 for three of four guidelines), while breadth counting was rule-sensitive: an explicit modifier-counting rule reduced the largest disagreement from 27 to 4 tokens. Residual disagreement was bounded and localizable to specific, nameable representational choices. Conclusions. Graph-extraction reproducibility can be systematically improved through iterative grammar refinement, and a prospectively derived rule can be confirmed to improve agreement. These results establish the measurement foundation (reliability, not construct validity) for a companion study interpreting CGCI as cognitive load, and the method may apply wherever graphs are extracted from structured source artifacts.