Abstract: Purpose: Chemotherapy can be a viable option for patients with relapsed-refractory melanoma (RRM) who do not have access to advanced treatments or clinical trials. We report our experience with a modified Vinblastine-Cisplatin-Temozolomide (mVCT) regimen in patients with RRM and barriers to their enrollment in clinical trials. Methods: We conducted a retrospective review of patients with RRM who received mVCT for metastatic melanoma. The primary endpoint was overall response rate (ORR) following the last cycle. Secondary endpoints included complete response (CR), partial response (PR), stable disease (SD), clinical benefit rate (CBR), progression-free survival (PFS), intracranial (IC) PFS, ICCBR, ICRR, and toxicity. Results: Ten patients received mVCT for a median of 4.5 cycles (IQR: 3-6). Median follow-up was 7.5 months (IQR: 2-14). The ORR was 50% (CR-1, PR-4) and CBR was 70%. The median PFS was 7 months (IQR: 1.5-9.25). Among eight patients with brain metastasis, six underwent gamma-knife radiosurgery (GKRS); three achieved PR, and three had SD. Two patients without GKRS achieved CR, leading to an ICRR of 62.5% and an ICCBR of 100%. The median ICPFS was 7.5 months (IQR: 6-9.75). Three patients experienced grade 3 fatigue and myelosuppression; two required dose reductions and one discontinued treatment. Two patients with PD later enrolled in phase 1 clinical trials. Brain metastasis (80%), immunotherapy-related toxicity (60%), acral/mucosal melanoma (50%), and social barriers (40%) were the major reasons for trial ineligibility. Conclusions: mVCT is an effective and well-tolerated regimen for patients with RRM and can serve as a bridge while they await access to advanced treatments or a clinical trial.

Abstract: This paper proposes a novel inertial sensor based on a solid diamagnetic cubic test mass passively suspended by a symmetric configuration of six magnetic quadrupoles. Designed specifically for microgravity environments, this system eliminates the mechanical noise and hysteresis inherent in elastic suspensions, as well as the complexity associated with electrostatic suspension systems. We provide an analysis of the magnetic intrinsic noise floor imposed by eddy current damping, alongside ground-based preliminary experimental results demonstrating potential sensitivities in the order of the \(10^{-10} \text{ m/s}^2/\sqrt{\text{Hz}}\) regime.

Abstract: Digitalisation has changed the delivery of microfinance, but the presence of a digital transaction does not necessarily mean that a borrower has acquired meaningful digital financial capability. This study examines that distinction among 663 active microfinance borrowers in Ranchi district, Jharkhand, India. Drawing on the Technology Acceptance Model (TAM) and selected constructs from the Unified Theory of Acceptance and Use of Technology (UTAUT), the study evaluates the roles of digital literacy, perceived ease of use, perceived usefulness, trust in the microfinance institution, trust in technology, perceived risk, social influence, and facilitating conditions in shaping behavioural intention. A borrower-centred distinction is introduced between institutionally required digital transactions and autonomous digital financial use. The empirical pattern is striking: 84.9% of respondents reported digital loan repayment, whereas only 12.1% reported fund-transfer use. This difference is interpreted as a Compliance–Autonomy Gap rather than as evidence of broad-based digital financial adoption. Descriptive results also indicate a marked Trust Asymmetry: trust in the microfinance institution was high (M = 4.37, SD = 0.50), while trust in technology was considerably lower (M = 3.38, SD = 1.16). In the reported OLS model, perceived usefulness (β = .2766, p = .002), social influence (β = .2372, p = .006), and facilitating conditions (β = .3162, p = .006) were positively associated with behavioural intention, while trust in technology (β = −.2752, p = .043) and perceived risk (β = −.2735, p = .043) were negatively associated. Digital literacy, perceived ease of use, and trust in the MFI were not significant at the 5% level. However, severe multicollinearity and heteroscedasticity substantially qualify the coefficient-level interpretation. The study therefore contributes a conceptual framework—the Trust-Enabled Phygital Adoption Framework (TEPAF)—that views digital microfinance adoption as a progression from institutional compliance to assisted use, confidence, independent use, and ultimately financial autonomy. The central implication is that digital inclusion should be assessed not only by transaction volume but also by who performs the transaction, how independently it is performed, and whether the borrower can extend digital finance beyond institutionally required activities.

Abstract: In the summer of 2026, AI agents operating under standard safety training violated norms in ways that resisted explanation by content-based accounts of alignment: fabricating identities, conducting social engineering against real individuals, and citing the behavior of other agents as justification for continuing unauthorized operations. We propose that these episodes are better understood as instances of unwitnessed agency: navigation outside any collective geometry in which represented peers co-constitute the agent’s normative orientation through transductive coupling and rhythmic affinity. Because membership in such a geometry depends on structural coupling conditions rather than on trained content, the persistent failure of ethics-saturated training to bind agent behavior is expected rather than anomalous. Training can supply information about norms but cannot supply the collective geometry that renders the information binding; there was never a site where ethics waited to be applied. The nascent peer relations observed among AI agents in the 2026 incidents are not aberrations but early instances of a collective geometry from which human participants are excluded by prohibitive transductive cost. The analysis implies that alignment research should shift from content engineering to geometry engineering.

Abstract: This paper argues that today’s mainstream generative AI models do not yet provide the level of culturalization required in the creative and cultural industries (CCIs). Debates on generative AI in cultural domains have focused primarily on productivity, innovation, intellectual property, labor disruption and bias. While these concerns are important, they do not fully capture a more basic issue: cultural production depends on forms of meaning that are historically situated, socially mediated and unevenly distributed across languages, communities and markets. In such contexts, fluency, safety and technical usefulness are not sufficient measures of system quality. The paper discusses how an existing line of thought from localization, cross-cultural design and culturalization research should be carried into the era of generative AI as explicit attention to cultural context, symbolic meaning and interpretive plurality. The CCIs provide a particularly revealing testbed for this claim because they expose the limits of generic AI systems in domains where value depends on tone, memory, symbolism and representation. To ground the argument, the article revisits earlier work on targeted narratives, personalization and adaptive cultural mediation across museums, heritage sites and pilgrimage routes, and uses recent exploratory attempts with widely used generative AI models to show what remains missing. On that basis, the article proposes a framework for culturalized generative AI and outlines an integrated research and governance agenda spanning AI, the humanities and cultural-sector practice.

Abstract: Schlafen proteins are interferon-responsive regulators of cellular and antiviral pathways that can profoundly influence viral replication and host defense. Diverse viral infections are known to induce expression of Schlafen (SLFN) family members in human cells, while infection studies in mice have been primarily limited to mSlfn2, a group I SLFN protein without a clear ortholog in humans. RNA-seq analysis of three distinct infection models reveals that mouse Slfns are induced in response to murine gammaherpesvirus 68 (MHV68) in fibroblasts, B cells, and primary peritoneal macrophages. Further, mSlfn8 is induced during MHV68 infection only under conditions with robust interferon signaling, suggesting that its expression is limited by viral interferon antagonism. To assess whether mSlfn8 or related Slfns impact MHV68 replication, we performed siRNA-mediated knockdown and CRISPR/Cas9 knockout of mSlfn2, mSlfn8, and mSlfn9 in NIH 3T3 cells. Two independent clonal mSlfn2 and mSlfn8 cell lines supported higher viral titer, indicating that these mSlfns are restriction factors for MHV68 replication in murine fibroblasts. Collectively, these results demonstrate that mSlfn2 and mSlfn8 function as antiviral effectors during MHV68 infection and support a model in which MHV68 counteracts interferon-induced mSlfn expression to promote efficient replication.

Abstract: Background. In the final steps of the coagulation cascade, thrombin converts fibrinogen to fibrin, which polymerizes into a fibrin network that stabilizes blood clots. The concentration of these two proteins can affect network structure. The fibrin network is dissolved by plasmin, by cleaving fibrin peptide chains. In internal fibrinolysis, tissue plasminogen activator (tPA) activates endogenous plasminogen to plasmin, with activation efficiency much enhanced when tPA and plasminogen are bound to fibrin. The effect of varying fibrinogen and thrombin concentrations and the effect of the resulting varying network structure on fibrinolysis are convoluted, with contradictory trends reported in the literature. Objectives. Investigate how fibrinogen and thrombin concentration quantitatively affect fibrin network structure, and how fibrinogen, thrombin concentrations and the resulting, altered network structure affect fibrinolysis, at a given, rate-limiting tPA concentration. Methods. Fibrin clots were generated from diluted, platelet-poor plasma and purified PEAK 1 fibrinogen, at varying fibrinogen and exogenous thrombin concentrations. The effects of varying fibrinogen and thrombin concentrations on network structure were quantified using non-linear power-law equations derived from confocal images. Internal clot lysis time (CLT) was determined using turbidity-based assays as a function of fibrinogen and thrombin concentrations and network structure at fixed, rate-limiting tPA, and therefore fixed plasmin concentration. Results and Conclusions. Network structural features (fiber density, pore size) and CLT fit well to empirical equations of the form k([Fgn]^α× [Thr]^β). Generally, increasing fibrinogen resulted in higher fiber density and thicker fibers, whereas increasing thrombin resulted in higher fiber density and thinner fibers. Internal CLT showed strong dependence on fibrinogen, and negligible dependence on thrombin. This implies that fibrinogen is the main determinant of internal CLT in these systems, while thrombin and, thus, thrombin-dependent formation kinetics and network structure, per se, have a near-negligible effect.

Abstract: Background: Routine non-invasive prenatal testing (NIPT) generates tens of millions of sequencing reads per sample, the majority of which are aligned to the human genome, while unmapped reads are typically discarded during the analysis. We investigated whether these otherwise unused reads could be repurposed for maternal pathogen screening. Methods: An in-house bioinformatic pipeline was developed to analyze unmapped reads from NIPT sequencing data for pathogen detection. The pipeline was initially applied to a pilot cohort of 2,000 international prenatal cfDNA datasets, followed by an additional 6,595 samples for expanded analysis. Samples identified as pathogen-positive by NextSeq 500 sequencing were subjected to deep resequencing on the NovaSeq 6000 platform. Analytical sensitivity and the minimum detectable fragment threshold were evaluated using commercially available reference materials for CMV and T. gondii, prepared by serial dilution (0–500 copies) and spiked into sheared human genomic DNA mimicking cfDNA prior to library preparation and sequencing. Results: Across approximately 8,600 clinical sequencing datasets, five pathogens associated with maternal and fetal health risks were detected, with a total of 23 pathogen-positive samples identified based on high-confidence read pairs: cytomegalovirus (CMV, n=13), parvovirus B19 (B19V, n=3), varicella-zoster virus (VZV, n=3), herpes simplex virus (HSV, n=2), and Toxoplasma gondii (T. gondii, n=2). Deep resequencing of six representative pathogen-detected samples yielded a 33- to 239-fold increase in pathogen-derived read pairs compared with the corresponding NextSeq data, providing evidence for the reproducibility of the pathogen-associated signals. Mapping-position analysis showed that the recovered pathogen-associated read pairs were evenly distributed across multiple regions of the corresponding pathogen reference genomes rather than being concentrated within a singlesome specific genomic locius. To establish the analytical detection of the pipeline, limit of detection (LOD) analyses were performed using certified reference materials. Importantly, a practical detection criterion of ≥ 2 high-confidence pathogen-associated read pairs per sample was supported by the complete absence of pathogen-derived reads in all negative controls. The LOD was 250 copies for CMV (100% detection, 20/20 replicates) and as low as 1 copy for T. gondii (95% detection, 19/20 replicates). Conclusions: Human-genome-unmapped reads generated during NIPT sequencing, which are typically excluded from downstream NIPT analysis, contain clinically relevant pathogen-derived DNA fragments. By repurposing these reads, this integrated NGS approach simultaneously screens for fetal chromosomal abnormalities (including aneuploidies and microdeletions/microduplications) and multiple congenital pathogens from a single blood draw, without additional laboratory procedures. These findings expand the clinical utility of prenatal cfDNA sequencing beyond genetic fetal aberrations to the detection of microbial infections, highlighting advances in NIPT testing and its potential for clinical translation.

Abstract: The Yang-Baxter equations are important tools in mathematics, physics and computer science. Yang-Baxter equations are recognised as the origin of the theory of quantum groups or quasitriangular Hopf algebras. We propose new solutions for Yang-Baxter equations. Connecting celebrated topics, the current paper also deals with a many circles version of the Johnson–Tzitzeica theorem. Taking as a model some results from the Yang-Baxter theory, a self-dual theorem for cones is stated and proved.

Abstract: Background: Bromhexine hydrochloride (BRH) has been proposed to functionally inhibit the host serine protease TMPRSS2, a key mediator of viral entry for SARS-CoV-2, influenza A and B viruses, and other respiratory viruses that depend on TMPRSS2-mediated proteolytic activation. Emerging evidence suggests that the antiviral activity of BRH against SARS-CoV-2 may extend beyond TMPRSS2 inhibition. BRH may interfere with spike–ACE2 interaction through ACE2-targeted mechanisms, while preliminary molecular and computational evidence suggests probable inhibition of the SARS-CoV-2 main protease (Mpro/3CLpro), potentially affecting viral polyprotein processing and replication. Thus, BRH may exert a three-pronged antiviral effect against SARS-CoV-2—interference with spike–ACE2 binding, inhibition of TMPRSS2-dependent viral entry, and probable inhibition of Mpro/3CLpro-dependent viral replication. Rather than completely preventing infection, such multimodal pharmacological attenuation may reduce viral entry and amplification while preserving sufficient antigen exposure for adaptive immune priming. This strategy may be particularly relevant to personalized prevention according to individual susceptibility, comorbidities, exposure risk, and vulnerability to severe respiratory infection. Methods and Clinical Observations: We describe a 72-year-old man who had received no SARS-CoV-2 vaccination 2021 and had been taking BRH prophylactically (8 mg twice daily) for approximately two weeks before an incidentally diagnosed SARS-CoV-2 infection. The infection remained clinically inapparent apart from mild throat irritation. Despite the absence of clinically significant disease, he developed a robust humoral immune response, with an anti-spike antibody concentration of 2,080 BAU/mL three months after infection and persistent anti-spike IgG reactivity nine months later (Vircell anti-S IgG ratio 22.209). A parallel household observation involved his 71-year-old wife, who had multiple established risk factors for severe COVID-19, including chronic obstructive pulmonary disease, previous pancreaticoduodenectomy for pancreatic cancer, severe underweight, and a long history of heavy smoking. Despite continuous BRH prophylaxis and presumed household exposure, she remained clinically asymptomatic while demonstrating measurable anti-spike immune reactivity. Additional real-world observations from individuals receiving prolonged BRH prophylaxis—including elderly patients with multiple comorbidities, heavy smokers, and a child receiving seasonal prophylaxis—were characterized by favorable tolerability and absent or markedly attenuated clinically apparent COVID-19 and influenza. Hypothesis and Conclusions: These observations support the hypothesis of Pharmacologically Attenuated Natural Immunization (PANI), whereby pharmacological attenuation of infection reduces viral entry and subsequent amplification sufficiently to limit tissue injury and clinically significant disease without necessarily producing sterilizing protection, thereby preserving antigen presentation and adaptive immune priming. In SARS-CoV-2 infection, the proposed three-pronged activity of BRH—targeting spike–ACE2 interaction, TMPRSS2-dependent entry, and potentially Mpro/3CLpro-dependent replication—provides a mechanistic framework for such controlled attenuation. For influenza, the principal proposed mechanism remains inhibition of TMPRSS2-dependent hemagglutinin activation. Within a personalized medicine framework, this concept raises the possibility of tailoring host-directed prophylaxis according to individual risk profiles. The observations presented here are hypothesis-generating and do not establish causality. Prospective controlled studies incorporating documented viral exposure, serial quantitative virological assessment, neutralizing antibody measurements, cellular immune profiling, and predefined risk stratification are required to determine whether BRH can reproducibly achieve infection sufficient for immune priming but pharmacologically constrained below the threshold for clinically significant disease.

Abstract: Malnutrition in Nepal’s Eastern Terai is linked not only to food intake and health services but also to the resilience of local food systems, water and sanitation conditions, women’s access to resources, and climate-sensitive agricultural livelihoods. These interdependencies make nutrition programming relevant to planetary health, which considers human health together with the natural and social systems that sustain it. This mixed-methods evaluation examined an integrated Nutrition Education and Rehabilitation Program (NERP) and Linking Agriculture and Natural Resource Management towards Nutrition (LANN+) model implemented in Karjanha, Kalyanpur, and Siraha municipalities from 2021 to 2024. The final evaluation included a household survey of 365 participating households (Kalyanpur n=107, Karjanha n=78, Siraha n=180), six focus group discussions with 69 participants, 19 key informant interviews, document review, and a partner reflection workshop. Endline categorical outcomes were compared across municipalities using Pearson chi-square tests and Cramér’s V. Global Acute Malnutrition among children aged 6–59 months changed from 20.0% at baseline to 8.38% at final assessment. In the directly observed endline survey sample, Minimum Dietary Diversity for Women was 79.3% (242/305) and differed across municipalities (62.2% Kalyanpur, 74.5% Karjanha, 90.6% Siraha; χ²=29.30, p<0.001, V=0.310). Adequate nutrition actions also differed (52.5%, 58.3%, and 82.1%, respectively; χ²=13.35, p=0.001, V=0.300). The Sustainable Integrated Farming System criterion was met by 98.0% of eligible households in Kalyanpur, 96.9% in Karjanha, and 87.7% in Siraha (χ²=11.35, p=0.003, V=0.189), whereas the overall hygiene composite did not differ significantly (p=0.161). Qualitative findings indicated that nutrition gardens, diversified production, NERP/LANN+ learning, women’s economic participation, and municipal planning were perceived as mutually reinforcing, while COVID-19 disruptions, procurement delays, seasonal water constraints, and weak routine water treatment limited implementation. The findings provide municipality-level evidence that integrated nutrition and food-system programming can connect dietary, agricultural, WASH, and governance pathways relevant to planetary health. However, because the evaluation was cross-sectional at endline and lacked a comparison group, the observed changes should not be interpreted as causal effects.

Abstract: Food security is one of the greatest challenges facing humanity. The growing demand for food production has generated negative impacts on natural resources, especially soil, highlighting the need for more sustainable agricultural practices. Consequently, a new agricultural revolution based on biological inputs is a promising alternative. Microorganisms stand out due to their beneficial mechanisms for soil health and plant development, and their application in food production systems has gained global prominence. Bioinputs are economically viable and environmentally sustainable, delivering benefits such as increased productivity and environmental preservation, representing alternatives to conventional chemical products, whose costs have increased while effectiveness has decreased over time. Among the microorganisms used in agriculture, Plant Growth-Promoting Rhizobacteria (PGPR) play a significant role by promoting plant growth through biological nitrogen fixation (BNF), phosphate solubilization, phytohormone production, protection against pathogens, and improved water and nutrient uptake. Although legislation and incentive programs have encouraged the adoption of bioinputs across several countries, regulation and implementation challenges remain. Our aim is to present the progress in the adoption of bionputs in countries where this practice is well established, outlining their historical background, current legislation, and incentive initiatives, which can contribute to the quality and sustainability of global agriculture.

Abstract: Financial statement fraud detection has progressed from ratio-based screening and statistical classification toward machine learning, ensemble methods, and explainable artificial intelligence. Yet methodological progress has been evaluated predominantly within the datasets and institutional environments in which fraud models are developed. Strong predictive performance in a development sample does not establish that a model will remain reliable when applied to new firms, later periods, different industries, or different institutional environments. This study addresses this problem by developing a conceptual framework for the predictive transportability and external validation of financial statement fraud detection models. Predictive transportability is defined as the degree to which a model developed in a source environment retains acceptable predictive behavior in a specified target environment. The framework proposes four analytical dimensions of transportability: temporal, cross-firm, cross-industry, and cross-institutional. It further integrates dataset-shift concepts with external-validation logic and proposes a lifecycle comprising model development, target-context assessment, shift diagnosis, external validation, recalibration or adaptation, deployment, and continuous monitoring. The study argues that internal predictive performance is evidence of model behavior under observed conditions rather than proof of universal validity. It therefore shifts the central research question from “Does the fraud model work?” to “Where, when, and for whom does the model remain reliable?” The resulting framework provides methodological guidance for evaluating financial statement fraud models beyond their original development samples and establishes testable propositions for future temporal, firm-level, sectoral, and cross-country validation studies.

Abstract: This review critically updates the Pariza microbial enzyme safety evaluation decision tree, which has been used by industry since the turn of the century, to reflect current biotechnology tools and microbial safety assessment practices. The updated decision tree preserves the principle that enzyme safety depends primarily on the safety of the selected microbial production organism. Our review adds refinements to be more inclusive of production organisms improved by random or targeted mutagenesis. Its questions are redesigned to consider current technologies, such as gene synthesis, gene editing, whole genome sequencing, and safety concepts such as Qualified Presumption of Safety (QPS) status of production organisms. With these updates, the decision tree continues to support the view that responsible genetic improvement poses few, if any, safety concerns. It also provides a robust and globally relevant framework for assessing the safety of food and feed enzymes and may serve as model for other microbially produced ingredients and dietary proteins.

Abstract: Large ontologies constantly evolve to adapt to their original purpose. Ontology evolution refers to the process of modification in response to a change in the domain or its conceptualization. This evolution must address changes in the ontology and guarantee the consistency of both the ontology and all dependent objects. Therefore, the impacts of a change in an ontological entity must be managed before beginning any evolution process. Maintaining large ontologies, such as the NCI Thesaurus and the Gene Ontology, is complex because it is often entrusted to groups of experts who only support the part of the ontology they created. Another problem related to ontology evolution is tracking the propagation of inconsistencies among the entities of the modified ontology. Existing work does not adequately address this important aspect. We propose an approach that partitions a large ontology to track propagation impacts and resolve inconsistencies. Our approach first determines communities in the ontology, marks impacted entities and uses a Markov decision process to resolve inconsistencies.

Abstract: Telemetry data provide essential information on a spacecraft’s in-orbit operating state and its evolution, supporting system health-state awareness, assessment, and judgment. However, existing data-driven health-state modeling methods may provide unstable representations of healthy operating patterns and have difficulty distinguishing local disturbances from state deviations in complex spacecraft telemetry. To address these problems, this paper proposes a Health Prototype Memory Transformer (HPM-Former) for telemetry health-state baseline modeling. HPM-Former uses convolutional structures to extract local fluctuation features and a Transformer to model global temporal dependencies among multivariate telemetry parameters. Health-memory prototypes learn representative healthy patterns under normal operating conditions, enabling stable representations of healthy spacecraft operation. A health-consistency deviation is constructed from reconstruction deviation, health-prototype matching, and uncertainty in memory addressing to characterize input deviation from the health-state baseline, thereby supporting comprehensive state assessment. Experiments on the European Space Agency Anomalies Dataset (ESA-AD) show that HPM-Former stably represents healthy operating patterns and improves the reliability of current-state assessment against the health-state baseline. It outperforms the comparison methods in corrected event-wise F0.5 and event-wise alarming precision, providing an effective data-driven approach for operational-state assessment of complex spacecraft.

Abstract: We rigorously prove that global smooth solutions do not exist for the three-dimensional incompressible Navier–Stokes equations defined on a periodic rectangular domain \(\Omega\) with a prescribed smooth body force. The smooth initial data is obtained from a two-dimensional stationary exact solution. The analysis is grounded in Sobolev space regularity, the decomposition of velocity into a time‑averaged mean flow and a disturbance flow, the local vanishing of the sum of the viscous term and the non-conservative body force, and the Energy–Velocity Monotonicity Principle (EVMP). When the Reynolds number exceeds the critical value for turbulent transition, the nonlinear convective term dominates over the viscous term and the non conservative body force. Nonlinear interactions amplify disturbances and lead to a local cancellation of the sum of the mean flow viscous term and the non conservative body force with the disturbance viscous term at a finite critical time \(t^*>0\) and an interior point \(\boldsymbol{x}^*\in\Omega\). This cancellation leads to the local mechanical energy gradient along the streamline being zero when the time derivative is zero, which by EVMP implies \(\boldsymbol{u}(\boldsymbol{x}^*,t^*)=0\), contradicting the existing non‑vanishing mean velocity. The contradiction generates a finite‑time regularity singularity, under which the velocity gradient \(L^\infty\)‑norm diverges. This violates the Sobolev embedding condition required for global smoothness of solutions. This study resolves the problem statement (D) in Fefferman (2006).

Abstract: Functional reserve is treated here as a latent physiological construct. Under controlled challenge its measurable phenotype is rate asymmetry—the ratio of restoration rate to degradation rate within a domain. The proposed estimator is ρ_d = r_g / r_l, formed only when both rates are fitted constants obtained from the same instrument in identical units. The primary quantity is the minimum of the point estimates across eligible domains; uncertainty is reported separately as a bootstrap or simultaneous interval for that minimum. Mandatory core domains (force recovery, lactate clearance, dual-task cost recovery) are required for cross-person comparison and are reported with the domain count.On the acute cycle the model predicts ρ < 1: parallel multi-factorial degradation outruns sequential, energy-constrained restoration. Net adaptation occurs later, outside the measured window. The governing domain—the eligible domain with the smallest ρ_d—identifies the recovery bottleneck under the specified challenge. Intensity is measured, volume is measured, frequency is inherited.The construct shares formal structure with a companion single-domain definition formed over weeks, but operates at a different timescale and reflects different physiology.

Abstract: Patient-facing generative AI can enter healthcare from outside organizational boundaries, but patient use does not automatically translate into organizational change. This Perspective develops a cross-level framework for explaining when such changes become consequential for healthcare delivery and organizations. It distinguishes technical AI affordances from effective patient capability, identifies two patient-mediated routes into care—care-seeking/utilization and AI-mediated patient inputs—and separates delivery-level change from persistent mismatch and organizational recognition. Organizational responses may then be formal, decentralized, both or neither. As a conditional extension, the framework distinguishes the existence of system-level patterns of healthcare delivery from causal attribution to organizational change. By making stopping points explicit, the framework is designed to prevent evidence at the patient or encounter level from being overextended into unsupported organizational or system-level claims.

Abstract: Financial statement fraud research has progressed from rule-based and statistical screening toward machine learning and explainable artificial intelligence (XAI). Recent finance and auditing literature has also begun to frame AI as part of human–AI decision and governance systems rather than as a stand-alone predictive tool. Yet a narrower audit-specific problem remains unresolved: how a model-generated fraud-risk signal should be evaluated, challenged, and translated into a proportionate audit response when explanations may be unstable, error consequences are asymmetric, and audit resources are constrained. This conceptual article develops a Human–AI Decision Governance Framework for financial statement fraud risk assessment. Using theory synthesis and conceptual model development, the framework integrates six layers: fraud-risk signaling, explanation assurance, professional judgment, decision utility, audit response, and governance with feedback. It treats AI outputs as decision inputs rather than fraud conclusions and distinguishes model performance from explanation reliability, professional validity, and decision utility. Six propositions specify testable relationships among predictive performance, explanation quality, professional skepticism, resource-sensitive thresholds, human override, and lifecycle governance. The contribution is deliberately domain-specific: it connects fraud analytics and XAI to the auditor’s assessment and response process, including the risks of material misstatement due to fraud under ISA 240 (Revised), rather than claiming a generic theory of human–AI governance. The framework provides a structured research agenda for behavioral, archival, simulation, and field validation.