Abstract: Morphogenetic System Theory (MST), proposed by Germano Resconi in 1998, is a generalized system model that integrates multidisciplinary knowledge and provides a unified mathematical framework for the recording, preservation, and reconstruction of forms. Although MST has been applied in risk analysis, system identification, and image recognition, its potential as a holistic meta-methodology for analyzing the evolution of socio-technical systems has not yet been fully revealed. This paper systemati-cally reviews the four-decade development of MST, distills its core philosophical propositions—the unity of subjective and objective understanding, the unity of analysis and synthesis, and the primacy of structure over individual values—and positions it as a holistic alternative to reductionist approaches. Systematic comparisons with seven related methods reveal MST’s unique capacity to capture both the formation and evolution of forms via the closed loop of write and read operations. Two case studies—learning community morphogenesis and legal morphogenesis—demonstrate MST’s applicability in socio-technical systems. The findings indicate that MST provides a formal, traceable, and scalable framework for quantitative morphological identification and evolution of socio-technical systems, with promise in education and institutional analysis. The paper also discusses the main limitations of MST and directions for future research.

Abstract: Bipolar membrane electrodialysis (BMED) is an electro-membrane separation technology used for over 20 years to treat liquid wastes, including wastewater from various industrial processes, municipal wastewater, and animal farms. This technology has been developed for its ability to dissociate H₂O into H+ and OH⁻ ions and to convert cations and anions in the feed solution into bases and acids, respectively, by applying an electrical potential. In comparison with alternative treatment technologies, BMED has some advantages, such as no need for chemical reagents, high degrees of efficiency, and straightforward operational procedures. However, BMED has been limited by a high level of preparation expense for BMs, membrane deposition, and ion loss during operation. This review mainly describes the principles and designs of BMED, their compositions along with their advantages and disadvantages, the membrane fouling and cleaning, and also the trend of their application in wastewater treatment. Finally, the results on nutrient recovery by BMED show that the performance of BMED depends on the nature of the ion exchange membranes used, the composition of the sample solution, the process conditions, and the configuration of the membrane stack.

Abstract: Transformative sustainability education often uses disruption to unsettle taken-for-granted assumptions, yet a generic ‘demands exceed resources’ account offers little theoretical leverage. This conceptual paper develops re-anchoring as a disruption-specific instructional design construct and reframes post-disruption pedagogy through three independently testable properties: resource sufficiency, functional congruence, and temporal sequencing. The framework predicts that, at equal support dose and scaffolding contingency, demand-congruent support will outperform incongruent support; that moderate congruent support may outperform higher-dose incongruent support; and that premature action support may reduce immediate distress while weakening delayed critical reflection, a false-resolution pattern for which moral licensing offers one candidate mechanism. Four anchors—meaning and values, responsibility and complexity, relational and future orientation, and feasible action—are derived from a post-disruption demand taxonomy, with GreenComp used only as a curriculum-translation vocabulary. Tourism education is theoretically diagnostic because sustainability critique there can implicate both consumer and emerging professional identities; consumer–professional identity overlap (CPIO) is therefore proposed as a conditional moderator for identity-implicating disruption only. Developed through integrative theory-building synthesis from a 50-source sensitizing corpus, the paper specifies rival-model tests, construct operationalizations, and five falsifiable propositions, together with the conditions under which the surrounding literature could be reconstructed to a standard permitting refutation.

Abstract: Nanomaterials such as fumed silica nanoparticles and nanocrystalline cellulose are emerging materials with many practical applications. One important application of these emerging materials is in the thickening and modification of the rheology of liquids. In this work, the rheological behavior of hybrid suspensions composed of fumed silica nanoparticles (T30) and nanocrystalline cellulose (NCC) was investigated to understand the combined effects of particle concentration and nanoparticle interactions on steady shear and viscoelastic properties. Steady shear measurements were conducted over a broad range of T30 concentrations, while oscillatory rheology was performed on selected high-silica formulations (7.06 and 8.1 wt% T30). All suspensions exhibited shear-thinning behavior, indicating progressive microstructural breakdown under increasing shear rate. The addition of NCC significantly increased the suspension consistency, suggesting enhanced particle–particle interactions and structural connectivity. Oscillatory measurements revealed elastic-dominated behavior (G'≫G'') across the entirely investigated frequency range for the selected formulations. The incorporation of NCC led to a substantial increase in both storage and loss moduli, with a more pronounced enhancement in the elastic response. In particular, the storage modulus increased by more than one order of magnitude (from ~210 Pa to ~8000 Pa at 1 rad/s), indicating the formation of a strongly interconnected viscoelastic network. The enhanced rheological behavior in hybrid suspensions is attributed to synergistic interactions between fumed silica aggregates and rod-like NCC particles, which promote network formation and mechanical reinforcement. Gap-dependent measurements showed negligible variation in the measured moduli, confirming the reliability of the rheological data. These findings demonstrate that small additions of NCC can transform fumed silica suspensions into strongly interconnected gel-like viscoelastic networks, which can be leveraged to enhance bulk rheological structuring and thickening of liquid systems.

Abstract: Labels are the bottleneck in underwater acoustic target recognition (UATR): hydrophones collect unlabeled audio essentially for free, while verified vessel labels remain scarce. Self-supervised learning (SSL) promises to convert the former into reusable representations, but its value for ship-radiated noise has been quantified only on single, leakage-prone corpora. We report a controlled three-corpus pilot: a ResNet-18 pretrained with SimCLR on 176,481 unlabeled log-mel segments pooled from the train splits of Oceanship, QiandaoEar22, and VTUAD, then evaluated on each corpus under linear probing and full fine-tuning against ImageNet and random initialization (15 corpus–initialization–protocol cells × three seeds: 45 runs). The result is predominantly negative at this budget. Under linear probing the SSL representation is significantly worse than ImageNet features on two of three corpora (VTUAD: −0.083 accuracy, p = 0.017; QiandaoEar22: −0.053, p = 0.012). Fine-tuning largely erases the differences within 30 epochs (paired accuracy gaps ≤ 0.039, mostly non-significant). One unestablished hint survives: QiandaoEar22 macro-F1 +0.0372 (p = 0.310). A diagnostic val–test gap on VTUAD, plus a recording-identity probe that reads recording ID near-perfectly from the SSL embeddings, shows the contrastive objective encodes recording identity, not vessel type. A normalization control rules out the main alternative explanation. We analyze why in-corpus contrastive pretraining under-delivers and derive requirements for the follow-on large-scale study. One limitation bounds all of this: pretraining was a single run (n = 1), so encoder-level variance is unmeasured, and all results are preliminary observations on one specific encoder, not general conclusions about SimCLR on this task.

Abstract: This paper presents a prospective multivariate analysis of Angola’s electricity sector from 2024 to 2050, assessing the economic, demographic, and environmental impacts of three energy transition pathways: Baseline (BS), Green Transition (GTS), and Restricted Development (RDS). Using energy system optimization models supported by parametric robustness tests, the analysis projects rapid electricity demand growth driven by population expansion to approximately 86 million inhabitants by 2050. Under the GTS scenario, electricity demand reaches 283 TWh, with per capita consumption increasing to 3,350 kWh/person. Although large-scale hydropower remains the backbone of the electricity system, supplying 37–51% of generation, solar photovoltaics become the dominant expansion technology, representing up to 33% of installed capacity while achieving the lowest Levelized Cost of Electricity (19.2–25.5 EUR/MWh). In contrast, diesel generation and new hydropower plants exhibit substantially higher generation costs. For isolated rural grids, diesel–solar hybrid systems reduce electricity production costs by more than 50% and significantly lower carbon intensity. Sensitivity analysis reveals that system costs are highly influenced by discount rate and macroeconomic assumptions but remain comparatively resilient to technological uncertainties. The findings demonstrate that large-scale solar deployment and decentralized hybrid systems are key strategies for achieving a cost-effective and low-carbon electricity transition in Angola.

Abstract: Visible–infrared detection improves perception in degraded illumination by combining visible texture with thermal contrast. Small objects remain difficult because their discriminative responses are weak, spatially sparse, and sensitive to cross-modal displacement. We present FGCFF-Net, a lightweight frequency-guided fusion detector for visible–infrared small object detection. The framework casts fusion as progressive evidence selection. A lightweight channel pruning module suppresses redundant dual-stream responses; frequency-guided alignment estimates residual spectral compensation; a cross-modal alignment and small-target framework predicts bidirectional spectral weights and deformable offsets; and a frequency-aware cross-modal aggregator conditions channel-spatial attention on frequency discrepancy. The training objective further couples dense detection loss with spectral consistency, offset smoothness, and small-target emphasis. On FLIR, M3FD, LLVIP, and KAIST, FGCFF-Netachieved 92.4% mAP50 on FLIR, 90.7% mAP50 on M3FD, 98.5% AP50 on LLVIP, and 5.72% log-average miss rate on KAIST, with 14.8M parameters and 68 FPS. These results indicate that frequency-aware alignment provides an efficient route for robust multispectral small-object perception.

Abstract: In Sityo Waldo, Tasyo is not simply believed; he becomes a test of whether a community can still ask questions when a speaker's identity begins to carry more weight than the reasons offered. Recent debates on moral testimony ask when deference is permissible or required; work on epistemic authority asks when testimony may rationally displace a hearer's own judgment; work on testimonial injustice examines how credibility assessments can wrong speakers; and work on normative powers and standing to blame shows that who speaks can sometimes partly determine what may validly be demanded. These debates share a question that is rarely stated in a common form: when a speaker's credibility, authority, or standing changes, what else changes with it? This essay offers a speaker-dependence test. It distinguishes the truth of a claim, the evidential role of a speaker, the independent normative force of a reason, and the relational force created by authority or standing. The central claim is that loss of trust in a person does not by itself defeat an independently grounded reason, although speaker identity can matter decisively when credibility is our evidence or when a speech act exercises a genuine normative power. Through the dispute between Tasyo and Mang Isko, and in conversation with current work on testimony, moral deference, epistemic injustice, authority, and blame, the essay shows why neither reverence nor contempt can replace the task of locating the source of a claim's force.

Abstract: The in-memory write buffer of an LSM-tree key-value store is searched on every read and written on every insert, so its cost per operation sets a floor on the whole engine. RocksDB uses a concurrent skip list, which chases pointers across independently allocated nodes and evaluates one data-dependent branch per key comparison. We present Aparajita, a MemTable representation that replaces the skip list with a list of cache-line-sized nodes, each holding fifteen 32-bit order-preserving key surrogates and a sentinel in one line, searched by a branchless SIMD kernel. Three design decisions carry the result. A relational vector compare over a sorted node yields a mask whose population count is the lower bound directly, so ordered search costs one compare, one movemask and one popcount with no branch. Surrogates are taken after the node’s shared prefix rather than from the start of the key, without which an absolute surrogate takes one value across all 200,000 keys in five of eight realistic distributions, including the keyspace this paper’s own evaluation runs on. Nodes are append-only, and the sorted order over their slots is a 64-bit word, so an insert is two stores into a free slot followed by one release store that publishes them. We implement Aparajita as a RocksDB plugin selectable by name without patching RocksDB sources, and evaluate it against the default skip list and VectorRep on a 12-core Emerald Rapids host. Point lookups over a resident MemTable are 22% to 39% faster than the skip list at 1, 4, 16 and 64 threads, non-overlapping across five runs per configuration at every point but one, backed by 55% fewer retired instructions and 40% fewer L1 misses per lookup. Ordered seeks are 15% to 29% faster, but a seek followed by ten iterator steps is 3% to 4% slower, and the representation charges 1.4 times the skip list’s arena per key. The skip list is 48% to 52% slower on insert at the representation level, and Aparajita is 18.8% faster in single-threaded db_bench, but the multi-threaded db_bench write path is bounded by RocksDB’s write group rather than by the MemTable: an insert there retires over 22,000 instructions in both representations. We report that ceiling rather than a write scaling claim the data does not support.

Abstract: Plant pangenomics has rapidly changed the way genetic diversity is represented and studied. During the last three years, advances in long-read sequencing, chromosome-scale genome assembly, structural-variant discovery, graph genomes, haplotype-resolved assemblies and super-pangenomes have demonstrated that a single reference genome is insufficient to describe the genetic diversity of crop species. Today the current challenges in plant pangenomics is the transformation of the pangenomics based information into gleanable resources, which could be biologically interpreted and useful for plant improvement. Pangenomics has been predominantly descriptive “catalogue of variation” paradigm and transformed toward a functional and predictive pangenomics framework integrating structural variation, regulatory sequences, transcriptomics, epigenomics, phenomics and artificial intelligence. This transition is particularly important for complex and polyploid crops, where conventional reference-based approaches remain inadequate. Super-pangenomes that incorporate wild relatives provide an important opportunity to recover genetic diversity lost during domestication. To achieve, pangenomes and super-pangenomes, computational scalability, functional validation, standardization and equitable access remain major challenges. In this paper, we present computational approaches to the pangenomics and introduce computational end to end scalabale pangenomic workflow and panscape, which serves a single standalone tool for all pangenomic needs.

Abstract: Mass polarization is often considered a partisan phenomenon, but is partisanship a necessary condition for mass polarization? To answer this question, we examine whether policy disagreement, specifically opinion divergence, can contribute to affective polarization independently of partisanship. We test three theoretical and empirical premises of partisan-centric accounts relating to opinion divergence. First, we test whether opinion divergence has remained stable, as previous conclusions of stability were instrumental in leading subsequent theories to de-center policy disagreement and prioritize partisanship. Using the Wasserstein Bipolarization Index that axiomatizes the distributional intuitions of bipolarization, we find increasing opinion divergence in the general electorate contrary to previous findings. Second, we test whether this increase extends beyond partisans by examining non-leaning Independents, whom partisan-centric models assume remain centrist over time. We find that non-leaners diverge comparably to weak partisans and leaning Independents, demonstrating that opinion divergence is not confined to the partisan framework. Third, to examine whether non-leaners' policy opinions have affective consequences, we test their association with affective differentiation based on agreement or disagreement and find a positive and growing correlation. Together, these findings provide evidence consistent with opinion divergence operating as a path to affective polarization independently of partisan identity. To reconcile this pathway with existing partisan-centric models, we propose a substantive-identity dual model of polarization.

Abstract: Poverty is closely associated with environmental conditions and access to essential environmental services, particularly in rapidly developing local communities. This study examines the relationship between environmental service conditions and poverty across 31 subdistricts in Bandung Regency, West Java, Indonesia, focusing on waste collection, access to safe drinking water, access to healthy sanitation, and slum settlements. The study employs a quantitative cross-sectional design using secondary subdistrict-level data for 2024. Multiple linear regression using the Ordinary Least Squares method was applied to examine the partial and joint relationships between the environmental indicators and poverty, complemented by diagnostic and sensitivity analyses. The results indicate that the environmental indicators are jointly associated with poverty, while waste collection is the only individual indicator showing a statistically significant negative relationship with poverty. Access to safe drinking water, healthy sanitation, and slum settlements do not exhibit statistically significant partial relationships with poverty. Sensitivity analysis further indicates that the negative relationship between waste collection and poverty remains stable after potentially influential observations are examined. These findings highlight the importance of incorporating environmental service provision, particularly waste collection, into locally targeted poverty reduction and public health strategies. Further research using longitudinal, household-level, and spatial data is needed to clarify the pathways linking environmental conditions and poverty.

Abstract: We report on our experimental study of polycrystalline Ge2Sb2Te5 including crystal structure (through x-ray diffraction and Raman spectroscopy), electrical resistivity ρ(T) (with applied magnetic field up to 35 Tesla), thermal conductivity κ(T), Seebeck coefficient S(T), and magnetic susceptibility in a wide temperature range. We find a residual resistivity ratio ρ(300 K)/ρ(1.8 K) of approximately 1.2 and a power-law, ρ(T) ∝ Tⁿ temperature dependent metallic behavior; where n ≈ 2 for1.8 K ≤ T ≤ 50 K and n ≈ 1.22 for 50 K < T ≤ 400 K. These results indicate that electron-phonon scattering dominates charge transport above 50 K, whereas electron-electron scattering becomes significant at lower temperatures. In applied magnetic fields B (up to 35 T) magnetoresistance is positive with an approximately linear √B dependence. Although at room temperature, thermal conductivity κ(T) is relatively low, about 2.5 W m⁻¹ K⁻¹ the Seebeck coefficient S(T) is also small. The implications of these findings are discussed.

Abstract: Memory-dependent evolution equations replace the time derivative of the state with a convolution against a memory kernel \( K(t) \), and appear throughout anomalous diffusion, viscoelasticity, and tumour-growth modelling.\( K \) is usually fixed a priori, most often to the power-law kernel of a Caputo fractional derivative. We investigate the joint recovery of the memory kernel K and the state u using a physics-informed neural network (PINN) trained by a block-coordinate Variable Projection (VarPro) scheme% , and we do so across four structurally different parametrizations: a non-negative Prony series, a fractional (Caputo) power law, a stretched exponential, and a shape-constrained free spline. On a synthetic benchmark the state is recovered with relative\( L^2 \) errors below \( 0.05\% \) across all four families, while kernel recovery varies sharply with the family: \( 0.03\% \) for the single-parameter fractional kernel, a \( 2--4\% \) structural gap for the Prony series that a systematic ablation (mode count, parameter bounds, data volume, optimizer iterations) fails to close, and an intermediate \( 5.7\% \) for the stretched exponential. A local sensitivity (Fisher-information-type) analysis at the exact solution, requiring no training, shows that the persistent instability of the individual Prony parameters is primarily associated with poor parameter identifiability rather than solely with optimization failure: its column-normalized condition number is more than two orders of magnitude worse when the true decay rates sit a factor of \( 4 \)apart than when they are well separated. Remarkably, this parameter-level non-identifiability does not prevent accurate recovery of the aggregated kernel function \( K(t) \) itself. For the stretched-exponential kernel, the initial instability was traced to a regularization term inherited from the Prony formulation. Removing this term reduced the kernel error from 13% to 5.7%. We further test and reject the classical hypothesis that a wider observation window would close the remaining gap. A shape-constrained spline that assumes only positivity, monotonicity, and convexity, with no exponential or power-law structure at all, recovers the benchmark kernel's shape to within \( 6\% \), matching the correctly specified parametric families and showing that the difficulty above sits in each family's own parametrization rather than in the learning problem itself. Finally, a Gauss--Newton/Laplace uncertainty analysis and an \( 8 \)-replicate nonlinear training ensemble converge on the same conclusion: even in the badly conditioned Prony case, individual rates carry uncertainty of hundreds to millions of percent, while the aggregated function \( K(t) \) stays determined to within a few percent, a quantitative instance of the classical inverse-problem distinction between parameter- and function-level identifiability. Overall, accurate state recovery is shown, both empirically and analytically, not to guarantee a uniquely recoverable memory operator.

Abstract: Recurrent landscape fires may threaten biodiversity in small protected areas embedded within human-dominated landscapes. We assessed fire occurrence and its spatial coincidence with species of conservation concern in protected natural area “Lalinačka Slatina”, by integrating stakeholder surveys, NASA FIRMS active-fire records, hydrometeorological data, and field-based mapping of selected plant and bird features. Stakeholders identified 48 fire-associated locations in the past decade, over 93% of which were linked to agricultural land or waste-disposal sites. The FIRMS dataset contained 1,073 active-fire detections, while 80% of MODIS detections recorded during 2001–2025 occurred between July and October. Maximum monthly temperature was positively associated with fire activity (IRR = 1.071, p = 0.0015), whereas precipitation had a negative effect (IRR = 0.957, p < 0.001). Direct correspondence between stakeholder-derived polygons and FIRMS detections was limited, indicating that the two sources captured complementary aspects of local fire occurrence. Fire exposure of selected species was spatially heterogeneous, reaching 98.8% in one mapped bird area and 100%, 96.5%, and 66.6% in several plant subpopulations. These results identify specific conservation-important areas where recurrent fire exposure is concentrated and provide a basis for targeted prevention, surveillance, and future ecological monitoring.

Abstract: Machine learning-based network intrusion detection, the explanation of detector decisions, and alert triage have each developed as a largely separate line of research. As a result, a detector’s aggregate performance score still tells a security analyst little about how far to trust an individual alert: it does not reveal how reliable the detector is for each attack type, whether the explanation of a decision is faithful, or on what calibrated basis some alerts should be automated and others escalated. We propose a confusion-aware triage framework that addresses these three concerns within a single, unified approach. From a trained detector’s predictions, the framework estimates the reliability of each attack type, validates the explanations of its decisions, and routes each alert, by that reliability, into automatic handling, analyst assistance, or escalation, presenting each escalated alert as a compact decision card. We evaluate the framework on UNSW-NB15 and CIC-IDS2017 with six detector families over five seeds. Per-attack-type reliability is estimable, stable across training runs, and transfers to unseen traffic, whereas a detector’s confidence overstates its reliability on the rarest attacks; the explanations are faithful for seven of the eight routed attack types; and routing by reliability escalates 27.9% of alerts while catching 75.7% of the detector’s errors, and abstains where the detector is already reliable. Routing by reliability matches routing by confidence on error coverage while remaining calibrated, per-attack-type, and auditable. The framework is proposed as an analyst-facing decision aid, and its evaluation with security analysts is left to future work.

Abstract: Background: Whole-blood transcriptomics captures systemic responses to infectious diarrhea, but pooled public-dataset reanalysis can be confounded by leukocyte composition and does not by itself establish cell-intrinsic regulation. Methods: We performed a profile-level secondary analysis of GSE69529 (255 profiles from 246 children; 203 diarrhea and 52 control profiles, including nine repeated profiles) using differential expression, functional enrichment, exploratory network analysis, promoter-sequence analysis, motif discovery, and transcription-factor activity inference. Because the pooled GEO2R contrast did not adjust for leukocyte composition, age, sex, pathogen, or repeated-subject structure, differentially expressed genes were interpreted as a composite whole-blood response. Published results from GSE276395 were used only for cross-study concordance, not as an independently reanalyzed validation dataset. Results: The pooled analysis identified 7,050 genes at adjusted P < 0.05 among 15,740 analyzed genes and highlighted broad immune, RNA-processing, cell-cycle, intracellular-organization, apoptosis, trafficking, and metabolic programs. Promoter composition was similar between regulatory directions; no HOCOMOCO motif was supported by AME, no STREME motif met the prespecified E-value threshold, and no CollecTRI-ULM transcription factor remained significant after false-discovery-rate correction. GSE276395 showed concordant innate-immune, neutrophil-centered, and immunometabolic themes, whereas its recurrent 12-gene set did not overlap the 12-gene discovery-stage panel. Conclusions: The reproducible signal is broad pathway-level whole-blood remodeling. Gene-, motif-, and regulator-level findings are exploratory, and the present design cannot distinguish shifts in leukocyte composition from within-cell transcriptional regulation.

Abstract: Pisolites are coated carbonate grains whose nucleus–cortex architecture may preserve evidence of changing hydrological conditions, yet their palaeoenvironmental significance remains poorly explored in Pleistocene Mediterranean coastal successions. This study investigates pisolite-bearing horizons from Eivissa (Balearic Islands, western Mediterranean) using field sedimentology, quantitative morphometry, petrography, and X-ray diffraction (XRD). The pisolites occur within fine-grained carbonate-cemented horizons associated with palaeosols and stabilized coastal dune systems, including interdune settings. Morphometric analysis of two closely spaced samples from the same Cala Compte horizon shows significant differences between matrix-associated and isolated fractions in maximum pisolite and nucleus diameters, while cortical thickness differs significantly only in CC2. Petrography reveals successive, locally discontinuous, asymmetric, and variably concentric cortexes, indicating polyphase accretion. XRD shows calcite-dominated pisolites and matrices with subordinate brushite, quartz, and minor unidentified phases. Integrated evidence supports local to near-local formation with limited redistribution under recurrent changes in water availability. The Eivissa pisolites therefore record episodic hydrological variability and polyphase carbonate formation during phases of Pleistocene coastal landscape stabilization.

Abstract: This paper investigates a nonlinear strong prescribed-time cooperative control theory, aiming to achieve spatio-temporal consistency in multi-agent control under disturbance. Taking multiple unmanned surface vessels(USVs) as the application object, a strong prescribed-time cooperative target-guided coordinate control(SPT-TACC) method is proposed to ensure that USVs can complete the cooperative target tracking task at the prescribed time. First, the error models of the cooperative kinematic subsystem and the kinetic subsystem are converted into periodic delayed forms, and a strong prescribed-time cooperative control system based on periodic delayed feedback is designed, enabling each USV to arrive at the desired tracking point simultaneously at the prescribed time. Second, a complete stability proof is provided for the nonlinear strong prescribed-time cooperative control theory with physical constraints. Finally, an empirical formula for the lower bound of the prescribed convergence time of the system with physical constraints is summarized, and the rules for setting the prescribed convergence time of each subsystem are summarized to avoid the divergence problem of the control system caused by unreasonable settings of multiple prescribed time. Simulation comparison results effectively verify the effectiveness of the proposed control method.

Abstract: Background: Hypoxemia after bidirectional Glenn palliation can result from several distinct physiologic mechanisms that may produce similar arterial oxygen saturation. We evaluated whether routinely available bedside markers can help distinguish these mechanisms in a mechanistic virtual-patient model.Methods: A steady-state zero-dimensional closed-loop model generated five causes of Glenn hypoxemia: elevated pulmonary vascular resistance, Veno venous collateral shunting, pulmonary gas-exchange impairment, increased metabolic demand, and ventricular dysfunction. The underlying physiologic variables responsible for each mechanism were hidden from the classifier. Machine-learning models used only bedside-observable variables: SpO₂, heart rate, mean arterial pressure, hemoglobin, Glenn/SVC pressure, and SVC oxygen saturation. Multinomial logistic regression was evaluated in a held-out test cohort, with random forest as a nonlinear comparator.Results: Among 150,000 generated states, 70,529 met prespecified hypoxemia and physiologic criteria; after class balancing, 26,905 virtual patients were analyzed. SpO₂ alone provided limited mechanistic discrimination (accuracy 26.8%; macro-AUROC 0.584). The multivariable bedside model improved accuracy to 59.1% (95% CI 57.9%–60.2%) and macro-AUROC to 0.865 (95% CI 0.860–0.871). Elevated pulmonary vascular resistance was most readily identified, with Glenn/SVC pressure providing the largest incremental contribution to classification. Veno venous collateral shunting and pulmonary gas-exchange impairment remained the principal areas of overlap. When compensatory systemic vasoconstriction was incorporated into ventricular dysfunction, systemic flow could fall despite preserved mean arterial pressure, reducing the discriminatory value of blood pressure alone.Conclusions: Bedside interpretation of Glenn hypoxemia should extend beyond arterial oxygen saturation. In this mechanistic model, Glenn/SVC pressure and venous oxygen saturation substantially improved distinction among physiologic causes of hypoxemia, while preserved arterial pressure did not reliably indicate preserved systemic flow. These findings support a physiology-based, multivariable approach to evaluating Glenn hypoxemia and provide specific hypotheses for prospective clinical validation.