Abstract: Establishing a rigorous nexus among reference-clock modeling, timing compensation, and system-level performance is imperative for cislunar positioning, navigation, and timing (PNT) architectures. In this study, we propose an integrated analytical framework to bridge relativistic clock definitions with PNT requirement allocations. Rather than treating timing budgets in isolation, the framework sequentially couples a first-order weak-field Earth–Moon clock difference model—which captures a temporal accumulation of approximately 20.48 ms annually—with chronological validations of empirical forecasting strategies. Furthermore, by propagating compensation residuals through a four-state nav-estimator, we demystify the assumption that residual timing translates linearly into position degradation. Crucially, we demonstrate that timing requirements in cislunar space are inextricably bound to source-dependent residual components, geometric dilution, and the overarching non-timing error floor, thereby mandating a joint geometric-stochastic budgeting paradigm for future lunar constellation designs.

Abstract: This article examines when a gender-responsive public procurement obligation becomes institutionally verifiable. It analyzes Brazil’s federal policy reserving jobs for women experiencing domestic and family violence, established by Law No. 14,133/2021 and regulated by Decrees No. 11,430/2023 and No. 12,516/2025. The study uses the empirical legal design documented in the underlying doctoral thesis and preserves four non-interchangeable evidentiary layers: an audit of 163 records; a retained corpus of 74 documentary units; a comparative matrix of 67 cases and 335 evidence records; and aggregate official and administrative information. The expanded matrix identified 16 cases with high density across all four dimensions, five with partial intersectoral integration, and 51 that were not verifiable through the documentary proxy. These figures describe the expanded corpus and are not used as a federal compliance denominator. Official aggregate data documented localized implementation involving 20 women; a later administrative update reported 143 women hired across five federative units, without contract-level granularity. The findings indicate that contractual enforceability, intersectoral connectivity, and protected verifiability are necessary institutional conditions—although not sufficient evidence of social impact—for converting the mandate into a monitorable practice. The article distinguishes direct federal evidence, normative diffusion, documentary absence, and non-verifiability without inferring a national effectiveness rate.

Abstract: Hundreds of survey studies have modelled the acceptance of generative artificial intelligence (GenAI) in higher education with partial least squares structural equation modeling (PLS-SEM), but their path coefficients have not been pooled, and it is not known whether the determinants established for earlier technologies hold, or hold evenly, for GenAI. We address this gap with the first meta-analysis of the 2026 PLS-SEM literature. Following PRISMA 2020, a Scopus search returned 151 records, of which 119 met strict eligibility. Standardized direct path coefficients and analytic sample sizes were extracted and double-checked by hand from each study's structural-model table. Six relationships with adequate coverage were pooled with random-effects models on Fisher-z-transformed coefficients (DerSimonian-Laird), and we report confidence and prediction intervals, subgroup and meta-regression moderators, study-level risk-of-bias appraisal, leave-one-out sensitivity analysis, and Egger's test with trim-and-fill. With between 11 and 20 studies per relationship and a combined sample of up to 10,819, all pooled effects were positive and significant: performance expectancy or usefulness predicted behavioural intention at beta = 0.33 (95% CI 0.25-0.41), effort expectancy or ease of use at 0.21, social influence at 0.21, and attitude at 0.58; behavioural intention predicted use at 0.52. Heterogeneity was very high throughout (I-squared 80-99%), and the prediction intervals were wide enough to include zero for four of the six relationships. A coherent regional pattern emerged: usefulness and effort expectancy were both weaker in Chinese samples (0.27 and 0.09) than elsewhere (0.37 and 0.29), whereas social influence was stronger (0.25 versus 0.17). Pooled estimates were robust to the removal of any single study; trim-and-fill imputed no missing studies. The acceptance hierarchy familiar from TAM and UTAUT is reproduced, but the determinants are far less stable than the pooled means suggest, and a single coefficient is a poor guide to any new setting.

Abstract: Background/Objectives: Since 2021, indeterminate thyroid nodules have been routinely assessed using molecular testing through Quebec’s provincial ThyroSeq v3 pilot project. The primary objective of our study was to compare the molecular and clinicopathological characteristics of Bethesda III and Bethesda IV thyroid nodules evaluated through the pilot project. The secondary objective was to assess the program’s economic impact. Methods: This retrospective study included 934 patients from two McGill University teaching hospitals in Quebec, comprising 593 Bethesda III and 341 Bethesda IV nodules that underwent molecular testing between 2021 and 2025. Eligible nodules were TI-RADS 3 or 4, measured 1–4 cm, and Bethesda III nodules required two consecutive cytology results before testing. Clinicopathological characteristics, histopathological diagnoses, molecular findings, and mutational profiles were analyzed. A projected 10-year cost analysis was also performed. Results: Of the 934 nodules tested, 280 demonstrated a genetic alteration. The risk of malignancy among surgically resected ThyroSeq v3-positive Bethesda III and IV nodules was 68% and 78%, respectively. Bethesda IV nodules demonstrated significantly greater vascular invasion (p = 0.029) and prevalence of oncocytic carcinomas (p = 0.010). Molecular testing was associated with an estimated CAD 1.49 million reduction in healthcare costs over 10 years. Conclusions: Although Bethesda IV nodules exhibited more oncocytic carcinomas and invasion, molecular mutation class distribution and the prevalence of aggressive pathological features were similar between Bethesda categories. Together with the demonstrated economic savings, these findings support continued implementation of publicly funded ThyroSeq v3 testing to improve management of indeterminate thyroid nodules while reducing unnecessary surgery.

Abstract: Purpose: The anaerobic speed reserve (ASR), defined as the difference between maximal sprint speed (MSS) and maximal aerobic speed (MAS), has been used in track and field to establish locomotor profiles and individualize training. In cross-country (XC) skiing, training encompasses varied activities that involve lower and upper body engagement and present distinct biomechanical and cardiometabolic characteristics. This study aimed to determine whether the locomotor profile of youth XC skiers depends on the locomotion mode. Methods: Twelve regional-level female skiers completed two field performance assessments in running (RUN) and roller skiing (SKI): a half-Cooper 6-minute test to determine MAS and a 40-m sprint to establish MSS. Three profiles (endurance, hybrid, or sprint) were determined from the ASR. Results: MAS was higher in SKI than RUN (17.2 ± 1.24 km·h⁻¹ vs 13.9 ± 1.0 km·h⁻¹; t (11) = 10.19, p < 0.001). MSS was also higher in SKI than RUN (23.0 ± 1.35 km·h⁻¹ vs 20.0 ± 1.3 km·h⁻¹; t (11) = 8.15; p < 0.001). However, the ASR did not significantly differ between modalities (p = 0.606). Most athletes were classified as hybrids in RUN (n = 7) and SKI (n = 6). Conclusion: Our findings indicate that most athletes retain a hybrid profile regardless of the modality, which is probably due to modern XC ski racing requirements. The stability of ASR across modalities suggests that this parameter reflects athlete’s intrinsic characteristics rather than modality-specific demands. The ASR may therefore represent a useful indicator for comparing athletes across different testing modalities.

Abstract: Background: The “fridge cigarette” is a social-media phenomenon in which a chilled diet carbonated soft drink (CSD) is ritualised as an adolescent stress-relief practice using the symbolic grammar of cigarette consumption. It presents an analytic problem for conventional prevention because its flagship product is sugar-free: harm is mediated not by sucrose but by acidity, non-nutritive sweetener exposure, and a transplanted dependency ritual. Linear exposure–outcome models do not account for the phenomenon’s propagation, its resistance to information-based correction, or the limited traction of existing regulation. Objective: To analyse the phenomenon as an instrumental case of a commercial determinant of health operating as a complex adaptive system, and to derive an implementable, theory-grounded intervention sited at high-leverage points in that system. Methods: A systems thinking review structured as a single instrumental case study. Evidence was assembled across five predefined system domains and synthesised using three established instruments: qualitative causal loop mapping with explicit link-polarity coding and evidence grading; systems-archetype analysis expressed as falsifiable propositions; and leverage-point classification following Meadows, in the shallow-to-deep formulation of Abson et al. Results: The case system comprises three reinforcing loops (R1 algorithmic amplification; R2 identity–affect reward; R3 commercial reinvestment) and one balancing loop (B1 clinical–regulatory correction). Loop-dominance analysis indicates that R1–R3 operate on cycle times of hours to months, whereas B1 carries a multi-year exposure-to-detection delay and low gain at two of three links, rendering it structurally incapable of stabilising the system. Four archetypes (Fixes that Fail, Shifting the Burden, Success to the Successful, Tragedy of the Commons) each predict that symptom-directed interventions will strengthen the reinforcing loops. Leverage classification shows the current policy repertoire concentrated in shallow tiers (L12–L6); no current instrument addresses the paradigm tier (L2–L1) at which the phenomenon operates. Conclusion: The analysis specifies CHAMP (Child-led Health Advocacy for Misinformation and Prevention), a peer-delivered counter-system of five sequenced levers mapped to deep leverage points, designed to re-aim the amplification loop rather than oppose it. The dental encounter is identified as the system’s earliest observable sensor, positioning pediatric dentistry to close the corrective feedback that currently fails. Recommendations are ranked by leverage depth, and an explicit falsification condition is stated.

Abstract: Olivine, as (Mg, Fe)2SiO4 solid solution, governs the plastic flow of the Earth's upper mantle. While extensive studies exist on natural olivine-rich rocks, the rheology of its Mg-end member, forsterite (Fo), remains less constrained, particularly for diffusion creep. Here we synthesize high-purity (≥98 vol.%), iron-free forsterite aggregates via pressureless sintering and perform axial compression experiments in a high-stress-precision Paterson gas-medium apparatus at 300 MPa, temperatures of 1423–1523 K, and differential stresses of 50–380 MPa. Our results reveal a stress exponent n = 1.0 ± 0.08, an activation energy Q = 365 ± 22.7 kJ/mol, and a grain size exponent p = 2.9 ± 0.23, demonstrating that forsterite deforms by diffusion creep under these conditions. The grain size exponent, close to the theoretical value of 3 for Coble creep, indicates that grain boundary diffusion is the rate-controlling mechanism. Compared to previous studies on forsterite and natural olivine, our flow law shows good agreement with the activation energy for olivine diffusion creep but provides a significantly better-constrained grain size exponent. Critically, because our samples are chemically synthesized and iron-free, and lack the trace impurities that facilitate defect generation in natural olivine, they exhibit higher strength than natural Fe-bearing olivine. Our flow law therefore defines the Mg-end member for the olivine solid solution system and represents the viscosity upper bound for olivine-dominated mantle rocks deformed dominantly by diffusion creep. These findings not only fill a critical gap in the rheological data for the olivine solid solution end-members but also provide a robust basis for modeling viscosity variations in the upper mantle as functions of grain size, temperature, and iron content.

Abstract: Predicting where a traffic agent will go and how it will get there are fundamentally different questions, yet most learning-based forecasters conflate them. When a model entangles agent intent (e.g., turning left) with driving style (e.g., cautious versus aggressive execution), spurious correlations arise that degrade predictions under distribution shift. This paper proposes SC-Mamba, a forecasting architecture built on selective state-space modeling and causal representation learning that explicitly disentangles these two factors. The model factorizes its latent space into independent intent and style variables, enforcing their separation through counterfactual do-interventions applied during training. A differentiable Signal Temporal Logic (STL) safety layer additionally constrains predicted trajectories to satisfy formal driving rules. Evaluated on the Argoverse 2 Motion Forecasting Dataset, SC-Mamba achieves 0.73 m minADE6, 1.22 m minFDE6, and 14.7% Miss Rate using only 1.35M parameters at 6.16 ms per scene—competitive with Transformer baselines at a fraction of computational cost. Critically, it is demonstrated that the learned intent latents correspond to identifiable maneuver categories (left turn, right turn, straight), style latents correlate with measurable execution dynamics (speed, curvature), and counterfactual manipulation of either factor independently controls trajectory shape without retraining. These results establish that linear-complexity SSM backbones can support structured causal reasoning for motion prediction while maintaining real-time performance.

Abstract: Availability-oriented monitoring may require low-latency anomaly inference, but deploying a lightweight model as a separate service adds process, serialization and network-path overhead. This article evaluates the inference-path architecture used by a prototype availability-response loop; it does not evaluate the effectiveness of a new DoS/EDoS detector or claim production attack mitigation. The same IsolationForest model and telemetry vectors were executed through an external Python/FastAPI sidecar and through ONNX Runtime embedded in a Spring Boot JVM. The prototype also included an LSTM Autoencoder and a bounded rule layer producing SCALE_UP, RETRY, FALLBACK or NONE. The main Kubernetes benchmark used Docker Desktop 4.73.1 (Docker Engine 29.4.3) with Kubernetes v1.34.3 on a single-node cluster. Embedded ONNX reduced request-response latency from 26.090 ms to 4.839 ms on average, from 62.065 ms to 5.367 ms at P95 and from 76.487 ms to 6.244 ms at P99; calculated sequential throughput increased from 38.3 to 206.7 requests/s. Additional checks covered concurrent load, gRPC transport, cold start, memory footprint and limited Kubernetes actions. In an extended comparison using 15 runs of 950 s for each active configuration, CPU-based HPA, rules-only and AI-plus-rules produced overlapping aggregate SLA-violation rates; no aggregate benefit of the ML gate over rules alone was observed under the tested degradation proxies. The evidence supports a narrow architectural conclusion: in-process ONNX is a lower-latency and lower-footprint execution path for the tested lightweight model. Detection quality, superiority over rule-only or established autoscaling mechanisms, and effectiveness against real adversarial traffic remain open validation tasks.

Abstract: The remarkable evolution of incretin-based therapeutics from selective glucagon-like peptide-1 receptor agonists to dual, triple and emerging polyagonists has fundamentally expanded our understanding of metabolic regulation. Clinical trials have consistently demonstrated coordinated improvements in glycaemic control, hepatic steatosis, body composition, insulin sensitivity, cardiovascular and renal outcomes, inflammatory activity, mitochondrial metabolism and energy expenditure that extend well beyond the expected physiological effects of individual receptor activation. These observations suggest that the therapeutic efficacy of contemporary polyagonists cannot be adequately explained by isolated hormonal mechanisms but instead reflects modulation of an integrated physiological regulatory system. We propose the existence of an Entero–Pancreatico–Hepatico–Systemic Polyhormonal Metabolic Homeostasis Network (EPHS-PMHN), a dynamic endocrine, neural, immune and metabolic communication network in which gastrointestinal nutrient sensing initiates coordinated signaling through the enteroendocrine system, pancreatic islets, liver, central nervous system, adipose tissue, skeletal muscle, kidneys, cardiovascular system, immune system and peripheral mitochondria. Although incretin hormones constitute the principal initiating signals within this network, physiological regulation is ultimately achieved through integrated actions of both incretin and non-incretin mediators including insulin, glucagon, amylin, fibroblast growth factor-21, bile acid signaling pathways, adipokines, hepatokines, myokines, autonomic neural pathways, inflammatory mediators, circadian regulators and intracellular nutrient-sensing mechanisms. Within this framework, physiological variables traditionally regarded as independently regulated — including fasting and postprandial glycaemia, hepatic fat content, insulin-glucagon balance, hepatic gluconeogenesis, lipid oxidation, triglyceride synthesis, resting energy expenditure, metabolic flexibility, mitochondrial function and inflammatory tone — are interpreted as emergent properties of coordinated network behaviour rather than isolated homeostatic endpoints. Furthermore, neurobehavioral determinants including psychological stress, sleep architecture, cognitive function, emotional state, personality traits and environmental influences continuously modulate network activity, thereby contributing to the unique metabolic phenotype observed in each individual. Accordingly, obesity, type 2 diabetes mellitus, metabolic dysfunction-associated steatotic liver disease, dyslipidaemia, sarcopenic obesity and cardio-renal-metabolic syndrome are viewed not as isolated disorders but as diverse clinical expressions of differential dysregulation within the same integrated metabolic network. This framework provides a mechanistic explanation for the superior efficacy of polyagonist therapies as network-restorative interventions rather than hormone replacement strategies and establishes a conceptual foundation for precision metabolic medicine based on characterization of network function instead of isolated biochemical abnormalities.

Abstract: Background: Surgical site infection (SSI) is a common complication after colorectal cancer (CRC) surgery. Although inflammation-based biomarkers have been investigated as prognostic indicators in CRC, their value for predicting superficial SSI using early postoperative laboratory findings remains unclear. We evaluated the utility of routinely available perioperative biomarkers for early prediction of superficial SSI after curative CRC surgery. Methods: This retrospective study included 488 patients who underwent elective curative resection for CRC. Preoperative and postoperative day (POD) 1 laboratory parameters, including serum albumin and inflammation-based biomarkers, were evaluated. Receiver operating characteristic curve analysis was performed to determine optimal cutoff values, followed by separate multivariable logistic regression analyses for preoperative and POD1 variables. Results: Superficial SSI developed in 66 patients (13.5%). A preoperative serum albumin level ≤3.70 g/dL was the only independent preoperative risk factor for superficial SSI (odds ratio [OR], 2.340; 95% confidence interval [CI], 1.166–4.695; P = 0.017). On POD1, a serum albumin level ≤2.80 g/dL (OR, 2.113; 95% CI, 1.021–4.376; P = 0.044) and a lymphocyte-to-monocyte ratio ≤4.60 (OR, 3.598; 95% CI, 1.559–8.302; P = 0.003) independently predicted superficial SSI. A combined model using preoperative serum albumin, POD1 serum albumin, and POD1 lymphocyte-to-monocyte ratio achieved an AUC of 0.691 (95% CI, 0.619–0.763), with 51.5% sensitivity and 80.3% specificity. Conclusions: Preoperative hypoalbuminemia and lower POD1 serum albumin and lymphocyte-to-monocyte ratio were independently associated with superficial SSI after curative CRC surgery. Assessment of these biomarkers may facilitate early identification of high-risk patients and support targeted preventive strategies.

Abstract:

Background: Ankylosing spondylitis (AS) associated with Crohn’s disease (CD) is a nosological form of spondyloarthropathies with a low population prevalence. The pathogenesis of this disease is not fully understood, and there are no precise differential diagnostic methods to distinguish AS associated with CD from AS and CD separately in the early stages of the disease. The main objective for this study is to define lymphocyte-mediated immunity in patients with AS associated with CD. Methods: For the pilot study, we recruited 4 groups: CD (n=16), AS+CD (n=13), AS (n=13) and healthy controls (HC, n=26). Immune phenotyping of peripheral blood lymphocytes via flow cytometry. Results: In the AS+CD group, circulating regulatory T cells (Treg) levels were increased compared to the other groups. The frequency of CD73+ Tregs within the effector memory (EM) compartment was also elevated in the AS+CD group relative to the AS and CD groups alone. At the same time, the EM Tcyt and EM Th populations were lower in the AS+CD group compared to the CD group. Tfh17 levels were lower in the AS+CD group than in the AS group and showed a positive correlation with Bm5 and switched memory B cells. Additionally, Tfh17 and switched memory B cells correlated negatively with acute-phase markers, whereas Tfh2 correlated positively with the BASDAI activity index. Conclusions: In patients with AS associated with CD, exhaustion of the regulatory compartment of adaptive immunity is seen, whereas the humoral component appears oriented towards resolving the inflammatory responses.

Abstract: Purpose: To review how standardized ultrasound-based risk stratification and artificial-intelligence (AI) governance have reshaped early diagnostic pathways for ovarian cancer. Materials and Methods: We performed a critical narrative review of milestones in transvaginal ultrasound, IOTA descriptors, ADNEX, O-RADS US/MRI, population screening trials and AI reporting/governance standards. Priority was given to evidence that used external validation, calibration or explicit links between risk categories and clinical management. Results: Population screening with CA-125 and ultrasound has not reduced mortality in average-risk women; therefore, the highest near-term value lies in triage once an adnexal lesion is detected. IOTA/ADNEX and O-RADS translate sonographic descriptors into reproducible probabilities and management categories; O-RADS MRI is useful for sonographically indeterminate lesions. AI may support segmentation, structured reporting and triage, but high discrimination is insufficient without external validation, calibration, transportability testing, decision-curve analysis and subgroup monitoring. Conclusion: Earlier and safer decisions are more likely to come from living standards - structured ultrasound reporting, calibrated risk models, MRI arbitration, auditable AI updates and equity-aware monitoring - than from population screening alone.

Abstract: Thermal error remains a primary bottleneck limiting the machining accuracy of preci-sion CNC machine tools. As a proactive, source-level countermeasure, thermal design has become increasingly critical for enabling next-generation high-performance ma-chine tools. This paper presents a critical and systematic review of machine-level thermal design methodologies, categorizing existing approaches into three principal technical routes: temperature control, material improvement, and structural optimiza-tion. For each route, we critically examine the underlying theoretical foundations, representative implementations, and reported effectiveness, with particular emphasis on the persistent gap between academic research and industrial practice. Critically, we find that existing thermal design efforts overwhelmingly target the magnitude of thermal deformation, while neglecting its spatial pattern, a deficiency that fundamen-tally undermines the effectiveness of subsequent thermal error compensation. In re-sponse to this critical gap, we argue that future thermal design should shift its para-digm from “amplitude minimization” to “deformation mode regularization”, actively shaping the spatial distribution and temporal evolution of thermal deformation to make it more predictable, repeatable, and readily compensable. This review concludes by outlining a forward-looking framework that integrates thermal mode analysis, dig-ital twin-based thermal state perception, and design-for-compensation principles, of-fering both theoretical foundations and practical guidelines for the thermal design of next-generation high-precision machine tools.

Abstract: Deep Reinforcement Learning (DRL) algorithms are frequently compared on standard bench mark environments, yet the practical implications of algorithmic choice for engineered, par tially observable tasks are less often examined side by side with such benchmarks. In this work we present a two-part empirical study. First, we implement ve DRL algorithms en tirely from scratch in PyTorch Deep Q-Network (DQN), Double DQN, REINFORCE, Proximal Policy Optimization (PPO), and Advantage Actor-Critic (A2C) and evaluate them under matched conditions on the LunarLander-v2 control task from the Gymnasium suite. We report learning curves, sample e ciency, stability, and nal policy quality, and nd that the value-based methods (DQN, Double DQN) solve the environment reliably (in 461 and 798 episodes respectively, reaching average scores above 200), whereas the on-policy methods REINFORCE, PPO, and A2C fail to solve the task within their allotted training budgets despite requiring substantially more environment interactions. Second, we design and evaluate Snow Drone Rescue, a custom, partially observable, stochastic environment built in pygame that models an autonomous drone performing a multi-objective search-and rescue mission under battery constraints and adverse weather. A DQN agent trained with a dense-plus-sparse shaped reward learns energy-aware navigation and obstacle avoidance despite receiving only a six-dimensional partial observation. Taken together, the two stud ies indicate that, for discrete-action tasks with reasonably dense reward signals, algorithmic stability mechanisms (experience replay, target networks) matter more for nal performance than the choice between value-based and policy-gradient families, while reward and envi ronment design remain at least as consequential as the learning algorithm itself. We discuss the theoretical basis for these observations, algorithmic and environment-design trade-o s, and directions for future work, including entropy-regularized policy optimization, systematic hyperparameter search, and extension to continuous-control and multi-agent settings.

Abstract: Plant viruses often depend on insect vectors for transmission, and interactions between viral coat proteins (CP) and insect endosymbiont-derived chaperonins may facilitate viral persistence within the vector. This study investigated interactions between a GroEL-homolog protein (GroHp) produced by the beet leafhopper endosymbiont Sulcia muelleri (SMBLH GroHp) and several plant virus CPs. Using PCR-capture assays, ELISA, and bacterial two-hybrid assays, we evaluated binding between SMBLH GroHp and multiple plant viruses and compared these interactions with those of Escherichia coli GroEL. Contrary to expectations, SMBLH GroHp did not exhibit strong specificity toward curtovirus CPs and showed weak interactions across multiple viral proteins, while GroEL displayed stronger and broader binding. Structural and biochemical characteristics of CPs, including charge and isoelectric point, were insufficient to predict binding affinity, suggesting additional molecular determinants influence GroHp-CP interactions. Furthermore, GroHp abundance did not differ between viruliferous and aviruliferous insects, indicating that GroHp is constitutively expressed and not induced by viral presence. These findings suggest that GroHp-CP interactions alone are unlikely to fully determine vector specificity or transmission efficiency. Instead, viral transmission likely depends on combined factors involving viral properties, vector physiology, and potentially additional molecular interactions within the vector.

Abstract: Behavioral control requires translation between different representations. Drives translate physiological state into motivational intensity. Values translate competing drives and context into an ordered ranking. Goals translate current values into temporally extended plans. Each translation is a protocol that transforms the representation of one domain into the representation of another domain. I argue that drives, values, and goals form a protocol stack of the kind found in many robust layered systems in engineering and biology. Each protocol occurs at a translation interface between domains. At the interface, heterogeneous upstream inputs compress into compact signals that downstream systems use to guide action. Interfaces tend to be evolutionarily conserved because change breaks communication between the two sides. Fragility concentrates at the interface. The framework predicts a gradient of model-dependence in the architecture of translation interfaces. Drives translate physiology into motivation through concentrated, identifiable interface circuits. Values combine concentrated components for direct translations and distributed components for model-based predictions. Goals primarily output model-based plans about extended futures, implemented in distributed circuits. Many observations match the framework's combination of translation, compression, conservation, and fragility. Examples from mammals include concentrated hypothalamic drive circuits, value interfaces combining concentrated midbrain dopaminergic and ventral striatal components with distributed orbitofrontal and ventromedial prefrontal cortex, and distributed goal circuits across prefrontal cortex and its recurrent loops with mediodorsal thalamus.

Abstract: Objective. Token-level confidence estimates from large language models (LLMs) do not reflect whether a generated clinical summary is factually correct. We investigate whether verification-based confidence prediction improves upon token-level estimates and whether LLM-as-judge labels can be assumed reliable for clinical summarization evaluation. Methods. Using MIMIC-IV, we construct 500 clinical summarization tasks across four open-source LLMs (Qwen2.5-14B/32B, Phi-4, Llama-3.1-8B), producing 2,000 primary summaries and 1,500 self-consistency samples. Each summary is scored on 26 verification features spanning coverage, factual consistency, natural language inference, numerical accuracy, semantic similarity, and self-consistency. Ground-truth labels are generated by Codex-GPT-5 and validated by two board-certified physicians on 200 summaries. Results. Three judges produce markedly different label distributions (Codex-GPT-5: 93.4% problematic, Phi-4: 10.5%, physicians: 59.5%) with near-zero pairwise agreement (\(\kappa = 0.009\)–$0.151$). Each physician independently identified only 1 of 200 summaries as fully trustworthy. ClinTrust achieves AUROC = 0.846 [95% CI: 0.788, 0.896] against Codex-GPT-5 labels and 0.695 [0.619, 0.768] against physician labels, significantly outperforming single-feature baselines including token log-probability (0.583) and BERTScore-F1 (0.607). Conclusions. Multi-source verification improves upon token-level confidence for predicting summary reliability, but LLM-as-judge labels are highly variable and cannot be assumed reliable without human validation.

Abstract: Background. The World Health Organization (WHO) Child Growth Standards provide a common language for screening and surveillance, but a z score is neither an etiologic diagnosis nor a complete account of a child’s growth trajectory. In Wayuu children, interpretation occurs amid documented malnutrition, food and water insecurity, territorial barriers, discontinuous care, and rights violations. This review examined whether current evidence supports claims of systematic anthropometric misclassification and developed an equity-oriented model for interpretation and action. Methods. We conducted a critical integrative review with structured, reproducible evidence identification completed on 04 January 2026. Direct quantitative and qualitative Wayuu evidence was analysed separately from comparative Indigenous evidence and methodological or normative sources. Bibliographic identity was verified against journal, DOI, PubMed, or institutional records. Findings were integrated through a proposition audit that distinguished directly supported conclusions, indirect plausibility, and unsupported claims. Meta-analysis and formal certainty grading were not undertaken because outcomes and designs were heterogeneous and no independent reference standard allowed estimation of diagnostic error. Results. Direct studies document a severe burden of chronic and acute nutritional compromise and multidimensional pathways involving food access, maternal-child nutrition, health-service contact, sanitation, mobility, income, language, and historical power relations. No direct evidence was identified for a Wayuu-specific mesomorphic somatotype in young children, a water-saving metabolic adaptation, or a quantifiable false-positive rate attributable to WHO standards. The main equity risk lies in the diagnostic pathway: harm can arise when measurement is unreliable, a screening classification substitutes for comprehensive assessment, relevant Wayuu knowledge is excluded, uncertainty is concealed, or decisions produce stigma, omission, delayed referral, or inequitable resource allocation. Conclusions. WHO standards should be retained for comparable screening and population monitoring, but interpreted through longitudinal, clinical, dietary, structural, linguistic, and community-informed assessment. Two safeguards are inseparable: do not pathologize difference, and do not naturalize preventable deprivation. Wayuu-led co-design, Indigenous data governance, outcome-linked validation, and equity audits are required before the proposed framework is used in policy or practice.

Abstract: Host immunological pathways can be categorized into IgM dominant innate immunological pathways, IgG dominant eradicable adaptive immunological pathways, and IgA dominant tolerable adaptive immunological pathways. The tolerable adaptive immune responses can be TH9, TH17, TH3, and TH4. Here, I further discuss the key cytokine in the TH4 immune response is interleukin-2. Once we know the whole framework of host immunological pathways, we can develop better prevention and control strategies combating infectious diseases and autoimmune disorders.