Abstract: Elucidating the pathophysiological mechanisms of mental disorders remains a critical challenge in psychiatric research. Recent studies have highlighted the potential involvement of cytoskeletal and molecular motor abnormalities in the development of mental disorders such as schizophrenia and autism spectrum disorder (ASD). This review synthesizes the latest findings on the relationship between cytoskeletal and molecular motor abnormalities and mental disorders. The cytoskeleton, composed of microtubules, actin filaments, and intermediate filaments, along with molecular motors such as kinesins, dyneins, and myosins, plays crucial roles in neurodevelopment, synapse formation, and neurotransmission. In schizophrenia, decreased expression of the microtubule-associated protein MAP2 and abnormalities in the DISC1 gene have been reported, potentially leading to dendritic morphological abnormalities and neurodevelopmental disorders. Additionally, abnormalities in molecular motors such as KIF17 and KIF1A have been implicated in disturbances of synaptic plasticity. In ASD, Myosin Id has been identified as a risk gene, and its localization in dendritic spines has recently been elucidated. Furthermore, abnormalities in actin-related proteins such as SHANK3 and CYFIP1 have been shown to cause synaptic dysfunction. These findings suggest that mental disorders arise from complex pathologies involving multiple cytoskeletal and molecular motor-related protein abnormalities. Future research should focus on elucidating the functions of individual proteins and adopting a comprehensive approach that includes glial cells. Advances in this field may deepen our understanding of the pathophysiological mechanisms of mental disorders and potentially lead to the development of novel therapeutic strategies.

Abstract: In classical mechanics, force is the physical entity mediating interactions between physical objects. Such objects consist of point masses, or appear as continuous bodies formed by a continuum of point masses. Force is defined as the sole entity capable of altering a point mass's state of motion (velocity) and is mathematically represented as a bound vector. However, this description of the physical world no longer holds at the atomic or subatomic level, where matter is discretized into quanta and interactions occur through the exchange of quanta of linear momentum and energy. While this dichotomy is currently accepted as the status quo, efforts to harmonize these frameworks into a more coherent formulation remain highly desirable. This paper investigates the extent to which interactions in classical mechanics can be reinterpreted as an exchange of linear momentum quanta. This investigation leads to a coherent reformulation of Newton’s laws, in which forces are treated as flow rates of these quanta. Therefore, classical mechanics admits a discretized description of the physical world even at the macroscopic level.

Abstract: Organizations operating in regulated industries must submit periodic reports to supervisory authorities. These reports often include financial disclosures, operational records, and compliance declarations. Traditional reporting methods involve manual document preparation and fragmented data sources. This research studies information systems designed for regulatory reporting and compliance documentation management. The proposed framework integrates structured reporting templates, data validation procedures, and submission tracking interfaces. System analysis demonstrates that centralized reporting platforms assist institutions in maintaining reporting accuracy and meeting regulatory submission deadlines.

Abstract: Tuberculosis (TB) is a subacute to chronic respiratory infection with insidious onset and protean symptoms. Treatment is complex and requires a multi-drug, multi-month regimen in which adherence is critical. Artificial intelligence (AI) offers promising solutions to challenges across the TB care cascade including screening, diagnosis and treatment. We conduct a scoping review of the literature published from 2017 to 2025 on the use of AI in TB screening, diagnosis, drug resistance diagnosis, treatment monitoring, and regimen design. We then extract data on study characteristics, AI methodology, input data, and sample size and describe AI tool performance and technology readiness level (TRL). Ninety studies are included, representing 803,383 study participants across 24 countries. Most studies (n=46) focus on radiological imaging for TB screening or diagnosis, but a burgeoning number of studies address drug resistance diagnosis (n=11), regimen design (n=4), treatment monitoring (n=12), and treatment adherence (n=8). Reported accuracy of AI interpretation of chest imaging for TB diagnosis was high at a median Area Under the receiver operating Curve (AUC) of 0.94 [IQR 0.12, range 0.81-0.99] for internal validation and 0.89 [IQR 0.14, range 0.66-0.98] for external validation, and a median TRL of 5 (IQR 1, range 4-7). AI demonstrates promise for advancing TB care towards World Health Organization (WHO) End TB targets, but several gaps remain, including AI-ready fit-for-task data availability, limited external validation and challenges in clinical integration. Closing these gaps will be critical for realizing the full potential of AI in TB care towards WHO End TB targets.

Abstract: Since the features of cross-seeding of alpha-synuclein forms may affect sensitivity and specificity of the test systems, we developed a modified approach to obtain alpha-synuclein amyloid seeds with particle sizes from 20 to 50 nm prepared from either the wild-type protein (α-synWT) or its more fibrillation-prone form A53T (α-synA53T). These seeds had optimal properties for subsequent initiation of fibrillation. Our data showed that the elevated efficiency of alpha-synuclein A53T monomers transformation was hardly affected by the type of used seeds, whereas the addition of the seeds obtained from the alpha-synuclein mutant form to wild-type protein monomers had a significantly less effect than α-synWT seeds. TEM data revealed that in the presence of α-synWT seeds the wild-type alpha-synuclein formed long and wide fibrils, while the addition of α-synA53T seeds led to the formation of long, but thin fibrils. The application of α-synA53T monomers significantly reduced the fibrillation lag period, making it a promising candidate for use in medical test systems. In the future, a set of alpha-synuclein mutant forms could be used for the differential diagnosis of synucleinopathies caused by the different mutations of this protein.

Abstract: This work explores the key capabilities of emerging sensing technologies in the context of Structural Health Monitoring (SHM) of civil infrastructures, aiming at contributing to the research on integrated and intelligent systems for more accessible and efficient monitoring solutions. As a case study, this study focuses on analysing the static and dynamic behaviour of the Edgar Cardoso stay-cable Bridge during its rehabilitation, recurring to a fully customized transducers and equipment. The developed system integrates sensors capable of measuring accelerations, displacements and temperature, which are connected to an autonomous data acquisition and transmission network. A digital interface was also developed to store, process and visualize the collected data, allowing remote access for later interpretation and analysis. The results confirmed the effectiveness of the developed system, which enabled the identification of the dynamic properties of the structure in terms of natural frequencies and vibration modes. The effects of traffic loads, as well as the correlations between temperature and structural displacements were also identified. Furthermore, the estimation of the axial forces in the stay cables permitted to study the influence of wind actions and traffic loads in these elements. The results demonstrate the potentialities of customized sensing solutions as effective tools for the management, maintenance, and long-term preservation of strategic infrastructures.

Abstract: This study examines the role of volunteers in the formation of social initiatives that utilize local resources. Previous research on volunteering has typically explained participation in terms of altruistic motives and a desire to contribute to society. However, this study focuses on intellectual curiosity—specifically, an interest in observing on-site situations and analyzing problem structures, as a factor that supports the continuity of volunteer activities.The study analyzes a local resource project that utilizes camellia leaves naturally growing on a remote island in Miyagi Prefecture, Japan. Using qualitative analysis, we examine the activities of a central practitioner, referred to as Practitioner A. Practitioner A plays a bridging role by connecting multiple actors, including local residents, companies, welfare facilities, tourism stakeholders, and researchers. This study conceptualizes such practitioners as “analytical volunteers.” Analytical volunteers are participants who are motivated not only by altruistic intentions but also by an intrinsic interest in observing real-world situations and constructing activities through problem analysis.The case analysis reveals that while the camellia project has succeeded in forming a network among diverse actors, it has not yet achieved stable commercialization. This stage is interpreted as the “growing pains” of social innovation in local communities.This study contributes to volunteer research by highlighting the presence of participants motivated by analytical thinking and by demonstrating the importance of network intermediaries in the formation of regional innovation processes.

Abstract: This research work deals with the challenges in software fault prediction (SFP) such as class imbalance in benchmark datasets, noisy features, and high-dimensional feature spaces. To overcome the above limitations, we propose a novel hybrid feature selection framework, FS-BWOA–COA, which incorporates Coati Optimization Algorithm (COA) for local exploitation and Beluga Whale Optimization Algorithm (BWOA) for global exploration. The two-phase optimization approach helps to avoid duplication and improves the stability of the classifier and also helps in maintaining the balance between exploration and exploitation. The framework was tested using several classifiers such as Decision Tree, SVM, KNN, and Naïve Bayes on eleven NASA PROMISE datasets. The hybrid outperforms single BWOA and COA, with an average accuracy of 0.9033 and peak values of 0.95 on the MC1 and JM1 datasets. The results of the statistical validation using the Friedman test, Wilcoxon signed-rank test, and paired t-tests confirm the same.

Abstract: Additive manufacturing is increasingly promoted as a pathway toward sustainable production; however, its Environmental, Social, and Governance (ESG) implications remain insufficiently defined at the operational level. This study addresses this gap by developing a governance implications assessment framework to interpret empirically validated life cycle sustainability assessment (LCSA) evidence for post-consumer recycled polylactic acid (PC-PLA)-based fused filament fabrication (FFF) within the Australian manufacturing context. The central contribution lies in demonstrating how LCSA evidence can be translated into decision-relevant ESG information without reliance on product declaration frameworks or simplified circularity claims. While LCAs and EPDs do not directly optimise production systems, sustainability-oriented governance can guide engineering and operational decisions by managing whole-system trade-offs and future risk. The framework integrates mechanical validation, service-life-normalised LCSA results, circularity metrics, and material flow modelling through a consequences analysis lens. Results indicate that a V50:R50 (vPLA:PC-PLA) blend achieves balanced performance, delivering a 57% reduction in global warming potential while maintaining functional durability. However, these benefits are governance-contingent; managing an 11.7% increase in service-life-normalised costs requires senior-level oversight, formalised material traceability, and structured workforce training. Waste diversion analysis further demonstrates that scalability is constrained primarily by upstream collection and sorting efficiency rather than fabrication performance. The study concludes with a strategic roadmap advocating regional recycling hubs, certified micro-credentials, and adoption of decision-relevant governance metrics aligned with ASRS S1 and S2 disclosure standards. The proposed framework offers a transferable approach for translating engineering-level sustainability evidence into credible governance insights, strengthening accountability and supporting Australia’s transition toward resilient, circular manufacturing.

Abstract: Motivation: Gene regulatory network (GRN) inference from single-cell RNA-seq (scRNA-seq) data remains hampered by technical noise, high false-positive rates, and extreme computational costs. Existing methods often require hours or days to process developmental datasets yet fail to capture the physical and topological constraints of regulatory interactions, essential for accurate regulatory mapping. Results: We developed AENetMoX, a multimodal autoencoder that integrates transcriptomic correlations with transcription factor (TF) binding motifs and protein-protein interaction (PPI) networks. We evaluated AENetMoX against SCENIC, GRNBoost2, and CLR across 48 independent configurations using three human brain organoid lineages. At K=100, AENetMoX achieved 7.7±5.1% precision (95.7% relative improvement over SCENIC; p=1.039×10^(-3), one-sided Wilcoxon test; Vargha-Delaney Â=0.635). ChIP-seq validation showed 51.6% precision (+9.5% over SCENIC, p=0.141, Â=0.500) and 168.4% improvement in F1 over SCENIC (p=2.883×10^(-9), Â=0.854). Ablation studies revealed PPI integration as the primary driver of performance, increasing precision by 492% (~5.9×) and ChIP-seq recovery precision by 198% (~3×) over its expression-only variant. Crucially, AENetMoX completes inference in under 5 minutes, a 24-fold speedup over SCENIC (~2 hours) and significantly outperforming CLR (>12 hours). Analysis of novel predictions identified 334 unique regulatory edges, including temporally persistent SMAD3 and SOX2 hubs that remain stable across multiple developmental stages. Availability: Source code is available at https://github.com/Shirshak52/AENetMoX. All datasets and databases used are available at OSF (Project DOI: https://doi.org/10.17605/OSF.IO/K6EHW).

Abstract: Thailand preserves one of the most extensive records of Mesozoic vertebrate footprints in Tropical Asia, yet these ichnological data have never been comprehensively synthesized. This review compiles and reassesses all known Triassic to Cretaceous vertebrate tracksites in Thailand to clarify their stratigraphic distribution, taxonomic diversity, and palaeobiogeographical significance. Published records, new field observations, and updated stratigraphic correlations are integrated to evaluate trackmaker attributions and temporal patterns. The Thai record documents diverse assemblages including chirotheriids, early theropods, sauropodomorphs, ornithopods, sauropods, and crocodilians. Upper Triassic–Lower Jurassic assemblages capture a major faunal transition, revealing the co-occurrence of non-dinosaurian archosaurs and some of the earliest dinosaurs in the region, whereas Lower Cretaceous sites are dominated by theropods, sauropods and diverse ornithopods. Comparison with other Asian ichnofaunas indicates faunal continuity across eastern Asia and supports early dinosaur dispersal into equatorial low latitudes. This synthesis also evaluates site conservation, highlighting the vulnerability of several Triassic localities and a positive trend of community-led discoveries since 2009, underscoring the need for proactive management and standardized digital documentation. Overall, the Thai ichnological succession represents the most complete Mesozoic footprint record presently known from Tropical Asia and provides key insights into vertebrate evolution, palaeoecology, and regional biogeography.

Abstract: Biometric identification has become a key element in modern security and surveillance applications; however, traditional systems based on a single biometric trait often suffer from noise, distortions, and vulnerability to manipulation, which limits their reliability in real-world environments. To overcome these challenges, this study proposes a multi-pattern biometric identification model that integrates facial features and hand gesture information extracted from video data for remote identity verification. The proposed system captures real-time video of an individual approaching a sensor, selects relevant frames, and applies advanced feature extraction techniques to both facial and hand modalities, which are then fused during the evaluation stage. Identity classification is performed using a time-delay neural network (TDNN), and the model is evaluated on diverse multimedia datasets containing static facial images and dynamic hand gestures, including American Sign Language samples. Experimental results demonstrate that the multimodal approach significantly outperforms single-modal systems, achieving an accuracy of 0.98, recall of 0.98, and F1 score of 0.97, compared to lower performance when using facial or hand features independently. These findings indicate that combining multiple biometric traits enhances robustness, reduces ambiguity, and improves recognition accuracy, making the proposed approach suitable for practical biometric verification scenarios under varying environmental conditions.

Abstract: Background: Access to prosthetic, orthotic and related assistive services remains uneven globally; this manuscript examines the systemic causes and rehabilitation consequences within the context of India. We frame service gaps as health-systems failures with measurable workforce, supply-chain, financing and data components. Methods: A narrative policy review was undertaken using targeted searches of peer-reviewed literature, government reports, professional body publications, and NGO datasets. Key themes were synthesized across governance, workforce, supply chain, financing, and monitoring domains to derive pragmatic policy interventions. (Authors should replace or update search dates and data sources as required prior to submission.) Findings/Observations: Four structural deficits drive undercoverage: (1) insufficient trained P&O workforce and uneven geographic distribution; (2) fragmented manufacturing and procurement with limited quality control; (3) inadequate public financing and poor insurance/benefits coverage for device services; and (4) absence of routine service and outcome surveillance. These deficits produce preventable functional dependency, increased caregiver burden, and inequitable access—most pronounced among rural, low-income, and disabled populations. Conclusions: Closing P&O service gaps requires integrated health-systems actions: workforce scale-up and credentialing, pooled procurement and quality standards, explicit public financing pathways, and routine service/outcome monitoring. Policy recommendations (summary): Five priority actions are proposed: national workforce strategy, accreditation and CE frameworks; standardized device procurement and quality assurance; finance and benefit design for assistive services; decentralized service hubs with tele-rehabilitation links; and a national monitoring dashboard tied to performance indicators.

Abstract:

Strawberry (Fragaria × ananassa Duch.) production faces growing pressure to reduce reliance on peat and coconut coir substrates, driven by documented life cycle liabilities including carbon losses from peat extraction and embodied transport emissions from coir. Nutrient film technique (NFT), a substrate-free recirculating hydroponic system, eliminates growing media entirely and reduces material inputs across successive crop cycles, making it an environmentally attractive candidate for controlled environment strawberry production. Despite early commercial adoption in Europe during the 1970s, NFT was largely abandoned for strawberry production by the 1980s following systematic failures whose physiological basis remains incompletely characterized. This review synthesizes evidence from hydroponic systems engineering, plant physiology, and oomycete pathology to examine the two structural constraints underlying NFT’s historical rejection: dissolved oxygen depletion dynamics within recirculating nutrient solution, and exceptional susceptibility to Pythium spp. root rot. We demonstrate that these constraints are coupled rather than independent, sharing a common pathway through root-zone oxygen status. Progressive root mat development over a six-month fruiting cycle degrades passive film aeration and creates hypoxic conditions that impair root membrane integrity, alter rhizosphere exudate profiles, and facilitate Pythium zoospore encystment and necrotrophic transition. This interaction is compounded by strawberry’s exceptional oxygen sensitivity and absence of adaptive aerenchyma formation, rendering thresholds established for tomato and cucumber inapplicable to this species. We identify two prerequisite research gaps that must be resolved before NFT can be rationally reconsidered for commercial strawberry production: characterization of root mat effects on channel hydraulic performance, and establishment of a strawberry-specific dissolved oxygen threshold under NFT-relevant conditions.

Abstract: The two spot cotton leafhopper (TSCL), Amrasca biguttula (Hemiptera: Cicadellidae), is an emerging invasive pest in the southeastern United States. Although TSCL is historically associated with cotton and vegetable crops, recent detections on ornamental hibiscus have raised regulatory concern, including “Stop Sale and Hold” orders and an emergency quarantine in Texas. Despite increasing pressure on hibiscus, no insecticide efficacy data exist for ornamental systems. We evaluated the acute (0–24 h) and residual (24–96 h) toxicity of bifenthrin, flupyradifurone, and tolfenpyrad against adult and immature TSCL using a sequential cohort leaf disc bioassay. New insects were introduced at 24 h and 72 h to isolate residue based mortality from prolonged exposure effects. Bifenthrin caused the highest acute mortality at 24 h, whereas flupyradifurone and tolfenpyrad exhibited slower initial activity but strong residual performance. Immatures were more susceptible than adults across doses. By 72 h, all three insecticides produced near complete mortality, with significant treatment and dose effects confirmed by ANOVA and binomial GLM analyses. Dose–response curves showed steep concentration dependent mortality for bifenthrin and tolfenpyrad and a time dependent response for flupyradifurone. These results provide the first insecticide efficacy data for TSCL on ornamental hibiscus and offer immediate guidance for nursery producers and regulatory agencies. The findings establish a foundation for whole plant and greenhouse evaluations to support integrated management and interstate plant movement compliance.

Abstract: The Gravity Recovery and Climate Experiment (GRACE/GRACE-FO) missions provide terrestrial water storage anomalies (TWSA) at coarse spatial resolution (300 km), limiting their application in medium-sized basins. This study develops a machine-learning framework to enhance the spatial interpretability of GRACE mascon TWSA within the 48,000 km² Hadejia-Jama’are River Basin, Nigeria. Hydroclimatic predictors derived from TerraClimate, Global Land Data Assimilation System (GLDAS), and Climate Hazards Group InfraRed Precipitation with Stations (CHIRPS) were integrated within a unified 4 km spatial framework. Four machine learning models were evaluated, including Random Forest (RF), Gradient Boosting, Histogram Gradient Boosting, and a Multi-Layer Perceptron. The RF model achieved the highest skill in reproducing mascon-scale TWSA (R² = 0.937; NSE = 0.937; RMSE = 4.36 cm). Aggregation of the 4 km fields back to the mascon scale preserved basin-integrated mass (R² = 0.94), confirming consistency with the original GRACE signal. The resulting groundwater storage anomaly (GWSA) fields resolve sub-basin spatial gradients and seasonal recharge-depletion cycles that are not discernible in the native product. Validation against 31 monitoring wells yielded moderate temporal agreement (Pearson correlation coefficient, r = 0.656), with magnitude discrepancies attributable primarily to scale mismatch and hydrogeological heterogeneity. While not a substitute for in-situ monitoring, the downscaled product enhances basin-scale groundwater assessment in data-scarce semi-arid regions. The framework is transferable to comparable basins and supports regional drought monitoring and water-resource management.

Abstract: This paper develops a hybrid quantum–classical framework for adaptive AI agents, combin- ing a self-reference-aware quantum evaluation layer with a classical candidate-generation and evolutionary optimization layer. On the quantum side, we introduce a nonlinear, memory- dependent extension of open-system dynamics through St[ρ] and derive key structural proper- ties, including trace preservation, Hermiticity, pointer-basis fixed-point behavior, and practical positivity conditions in bounded-coupling regimes. On the AI-systems side, we define measur- able response metrics (χ2, ζ), introduce a compositional synergy integral Sint, and specify an online-selection plus offline-evolution pipeline. Candidate-dependent evaluation is implemented through semantic embedding and amplitude encoding, so quantum initialization reflects linguis- tic proximity rather than hash collisions. The contribution is framed as a testable theoretical architecture rather than a universal performance claim: χ2 and ζ are structural diagnostics, while semantic-quality gains remain an empirical hypothesis requiring calibration. We also pro- vide implementation-oriented interfaces and a worked compositional example to support staged empirical validation on NISQ-era hardware.

Abstract: Contemporary theories of emotion explain affective life through biologically prepared modules, cognitive appraisal mechanisms, predictive construction, dimensional organization, neural survival circuits, attachment regulation, adaptive computation, and error minimization. While each framework offers important insight, none fully explains why emotional intensity varies dramatically across contexts, why identical informational input yields divergent responses, or why certain commitments reorganize identity and override instrumental reasoning.This paper introduces the Dynamic Love-Based Valuation (DLBV) framework, proposing that emotional diversity emerges from a unified valuation architecture grounded in identity-relevant valuation, termed love. Love is defined not as romantic behavior nor as a discrete emotion, but as a structural valuation system operating across three qualitatively distinct phases: attraction, immersion, and union. Attraction preserves instrumental rationality and goal-directed evaluation. Immersion reorganizes valuation through identity expansion and potential subordination of rational constraint. Union stabilizes valuation within integrated attachment and responsibility.Within this framework, emotions such as fear, anger, sadness, joy, jealousy, guilt, gratitude, hope, and despair are interpreted as contextual modulations of valuation under specific informational conditions, including threat, violation, deprivation, alignment, rivalry, uncertainty, or loss. Emotional tone and intensity depend not solely on appraisal content or arousal magnitude but on the structural depth at which valuation operates.To move beyond conceptual synthesis, the paper proposes an experimental paradigm designed to test phase-dependent modulation of emotional intensity. The design operationalizes identity-relevant valuation across the three love phases and examines whether identical informational stimuli elicit systematically different affective responses depending on phase-structured commitment. The Dynamic Love-Based Valuation framework therefore offers both an integrative theoretical account and a falsifiable empirical program for investigating the structural architecture of emotion.

Abstract: We report the first complete circular genome of Acetobacter cerevisiae KSO5, an indigenous strain isolated from Korean fruit vinegar, comprising a 3.3 Mb chromosome and two plasmids encoding 2,898 genes. Phylogenomics confirmed species assignment (average nucleotide identity, ANI 97%; digital DNA–DNA hybridization, dDDH 71%). Comparison with seven draft A. cerevisiae genomes revealed strain-specific genomic islands, mobile genetic elements and polymorphisms in stress-response pathways, with enrichment in acid-tolerance–associated functions, and highlighted plasmid-borne modules potentially linked to genetic stability. The genome encodes a periplasmic oxidative fermentation system with membrane-bound pyrroloquinoline quinone-dependent alcohol dehydrogenase (PQQ-ADH) and molybdopterin-dependent aldehyde dehydrogenase (Mo-ALDH), together with respiratory-chain components consistent with flexible aerobic metabolism. Three acetate-handling routes (efflux, acetyl-CoA conversion and an AarC branch) were also predicted, suggesting mechanisms to limit intracellular acetate accumulation. Consistent with these features, phenotyping under ethanol stress (5–10%) showed measurable growth and titratable acidity production up to 9% ethanol (late-stage peak acidity). These data provide a genomic and phenotypic basis for developing robust vinegar starter cultures.

Abstract: Background: Women with atrial fibrillation experience a higher lifetime risk of ischemic stroke, greater stroke severity, and worse functional outcomes than men. Preventive strategies focused on AF detection may therefore miss critical opportunities for early intervention in women; (2) Methods: We developed a decision-analytic Markov model using real-world primary care data from Catalonia (Spain) to evaluate an artificial intelligence (AI) enabled strategy for upstream thromboembolic risk detection. The intervention combined electronic health record–based risk prediction, targeted digital rhythm screening, and individualized anticoagulation. Lifetime clinical and economic outcomes were estimated for adults aged ≥65 years, with pre-specified sex-stratified analysis; (3) Results: Compared with usual care, the AI-enabled strategy reduced ischemic stroke, major adverse cardiovascular events, and long-term disability. Absolute reductions in stroke and disability were greater in women, reflecting higher baseline thromboembolic risk. Per 1,000 high-risk women, the strategy prevented more strokes and generated larger quality-adjusted life-year gains than in men. From both healthcare payer and societal perspectives, the intervention was cost-saving in women, driven by reductions in stroke-related disability and long-term care; (4) Conclusions: AI-enabled upstream thromboembolic risk detection may deliver particularly important benefits for older women and represents a promising approach to reduce sex-based inequities in stroke prevention.