Search Results (576)

Background/Objectives: Artificial intelligence (AI) models are increasingly applied across the assisted reproductive technology (ART) workflow, including male-factor assessment, ovarian stimulation, endometrial receptivity evaluation, embryo selection and prediction of pregnancy outcomes. However, many systems remain difficult to interpret, raising concerns regarding transparency, clinical integration and patient communication. Explainable artificial intelligence (XAI) aims to address these limitations by making model behavior more accessible to clinicians and embryologists. This review aimed to provide a narrative, concept-driven synthesis of how XAI has been implemented in ART, to critically examine methodological quality and clinical relevance and to outline priorities for responsible translation into practice. Methods: A structured narrative review was conducted using PubMed/MEDLINE as the primary database, supplemented by targeted reference-list screening of key primary studies and recent cross-disciplinary reviews relevant to AI in ART. Studies were curated and classified according to stage of the ART workflow, data modality, model family, explanation technique and validation strategy. Methodological features, performance reporting and implementation considerations were qualitatively appraised. Results: Most XAI applications in ART fall into two dominant categories: (i) feature-attribution methods such as SHAP and LIME applied to tabular clinical and laboratory data and (ii) saliency-based approaches, including Grad-CAM and related techniques, applied to embryo and ultrasound imaging. These methods can improve transparency and support counselling by clarifying which variables or image regions influence predictions. However, the majority of studies are retrospective and single centre, with limited external validation and heterogeneous outcome definitions, often prioritising clinical pregnancy over live birth. Calibration, decision-analytic evaluation and prospective assessment remain uncommon. XAI outputs are frequently interpreted as biologically causal despite being derived from observational data, highlighting the need for cautious clinical framing. Conclusions: XAI in ART has progressed from proof-of-concept demonstrations to early clinically oriented tools, but robust validation, standardised reporting and thoughtful workflow integration are still needed. Explanations can enhance auditability and communication, yet they do not compensate for methodological weakness. Future progress will depend on higher-quality multi-centre data, evaluation beyond discrimination metrics and governance frameworks that ensure transparency, fairness and sustained performance in real-world practice. Full article

Plant electrophysiological signals can rapidly reflect the dynamic responses of plants to external stimuli, giving them strong potential for nondestructive monitoring and early state recognition. However, differences among plant organs, as well as spatial heterogeneity within the same organ, may substantially affect signal quality and stability because of variations in tissue structure and local physiological activity. To address this issue, this study used Clivia as an experimental model and established a controlled local leaf-surface salt-treatment paradigm to systematically evaluate the relative discriminative ability of electrophysiological signals recorded from different spatial positions on leaves. First, stepwise screening of longitudinal leaf regions and leaf hierarchy was performed using 0 mM and 100 mM NaCl agarose gel treatments to determine the optimal signal acquisition position. Then, based on the selected position, a five-level NaCl treatment recognition task was constructed, and LRPNet, a residual network integrating PoolFormer and an efficient channel attention mechanism, was proposed for multi-gradient classification of plant electrophysiological signals. The results showed that, within the current experimental framework, the basal region of the top leaf exhibited the highest relative separability and the best overall recognition performance. In the five-gradient recognition task, LRPNet achieved the highest mean Accuracy of 92.21% among the compared models. These findings indicated that plant electrophysiological signals exhibited pronounced spatial heterogeneity and that optimization of the recording location was not merely an experimental detail, but an important upstream factor that affected downstream recognition performance. This study provides a methodological basis for optimizing signal acquisition positions and improving electrophysiological signal recognition in plants. However, the present conclusions are mainly applicable to the controlled local salt-treatment paradigm established in this study and still require further validation through more rigorous physiological verification, cross-scenario testing, and more independent data-partitioning strategies. Full article

Stepfamily formation involving adolescents presents unique challenges and opportunities, yet factors that shape stepparent–adolescent interaction remain under-explored. This study explored factors that facilitate or obstruct stepparent–adolescent interaction of any kind. Using maximum variation purposive sampling, we conducted in-depth interviews with 18 U.S. emerging adults (ages 18–23) who had lived with a stepparent from age 10 onward. Transcripts were analyzed using reflexive thematic analysis. Adapting the COM-B (capability, opportunity, motivation—behavior) model as a framework, 11 themes were developed: under capability—individual traits (with three subthemes), stepparent allegiance or deference to the resident parent, adolescent loyalty binds, relational ambiguity, and assumed roles and imposed expectations; under opportunity—external social facilitation (with three subthemes), stepparent compensatory and value-add functions, physical proximity and household composition, and shared family labor; and under motivation—adolescent interest in relationship development and stepparent proaction. Findings showcase stepparent–adolescent interaction as an emergent systemic property shaped by intrapersonal, dyadic, household, and broader social factors. The COM-B framework offers professionals a structured approach to assess stepparent–adolescent interactional capability, opportunity, and motivation when supporting stepfamilies. Future research should employ prospective, multi-informant designs to further substantiate identified factors and guide the development of measures for practice and research applications. Full article

The natural gas hydrate production riser is the main passage for offshore hydrate production and transport. Its safe operation directly affects the production process. However, current hydrate production methods cannot avoid hydrate decomposition and formation inside the pipe. Hydrate phase change causes internal multiphase flow. Together with the external ocean current, it leads to more complex nonlinear vibration of the riser. Based on China’s gas hydrate trial production in the Shenhu area of the South China Sea, this study establishes a dynamic model of a production riser. The model considers hydrate phase change inside the pipe and vortex-induced vibration. It is solved using the Newmark-β method, and its validity is confirmed by CFD simulations. The results show that, under the combined action of ocean currents and internal multiphase flow, the riser exhibits a clear multi-frequency response in vortex-induced vibration. Its spatial trajectory is highly irregular. Specifically, hydrate phase change increases internal gas content and gas slippage, elevating fluid velocity. This reduces the riser’s structural stiffness and effective tension, altering the VIV response. In addition, lower top tension and higher slurry density, flow rate, and outlet backpressure delay hydrate decomposition. These factors also reduce the effective tension along the riser and increase its in-line deformation. Full article

Mesenchymal stromal/stem cells (MSCs) are increasingly explored for immune-mediated diseases, yet standardized analytical readouts that capture coordinated immunomodulatory output across complementary secretory pathways remain limited. Here, we report the feasibility of an HPLC-based multi-analyte secretome characterization panel that quantifies two small-molecule outputs—adenosine and kynurenine—alongside two immunomodulatory proteins—interleukin-10 (IL-10) and transforming growth factor-beta (TGF-β)—in conditioned media from canine adipose-derived MSCs (cAD-MSCs). Canine immune-mediated hemolytic anemia (IMHA) was used as a disease context to motivate the selection of these analytes, given the pro-inflammatory cytokine environment characteristic of this condition. Three independent cAD-MSC lines were evaluated under baseline conditions and following cytokine stimulation with recombinant interferon-gamma (IFN-γ; 100 ng/mL) and tumor necrosis factor-alpha (TNF-α; 50 ng/mL), referred to herein as inflammatory priming or licensing. Conditioned media were collected at 72 h for metabolite analysis and 48 h for protein analysis, and quantified by HPLC using external calibration and peak integration. Across all three lines, licensing produced directionally consistent increases: mean adenosine increased 2.3-fold, mean kynurenine increased 3.1-fold, mean IL-10 increased 1.6-fold, and mean TGF-β increased 1.7-fold compared with unlicensed controls. Metabolite measurements for adenosine and kynurenine are reported with full chromatographic selectivity data; IL-10 and TGF-β measurements by reversed-phase HPLC with UV detection are presented as exploratory/semi-quantitative outputs and will require orthogonal confirmation (e.g., immunoassay) in future work. These findings are preliminary, derived from three independent donor lines with no comparator group, and are intended to support feasibility of the analytical framework rather than establish definitive performance specifications. Collectively, the data support the potential of a multi-analyte HPLC-based characterization panel to capture licensing-responsive secretory shifts across mechanistically complementary pathways, providing a foundation for expanded development and validation. Full article

Sustainable precision agriculture is crucial for optimizing resource utilization, reducing chemical inputs, and ensuring global food security. High-precision automatic recognition and monitoring of key crop organs (e.g., wheat heads and flower clusters) serve as the technological foundation for sustainable agricultural management decisions. However, visual perception in natural field environments is highly susceptible to external conditions. To address the challenges of severe background interference and feature dilution in crop small object detection within complex agricultural scenarios, this paper proposes an enhanced detection network, ACF-YOLO, based on YOLO11. First, an Aggregated Multi-scale Local-Global Attention (AMLGA) module is designed to enhance the feature representation of weak targets by fusing local details with global semantics. Second, a Context-Guided Fusion Module (CGFM) and a Soft-Neighbor Interpolation (SNI) strategy are introduced. Their synergy alleviates feature aliasing effects and ensures the precise alignment of deep semantic information with shallow spatial details. Furthermore, the Inner-MPDIoU loss function is employed to optimize the bounding box regression accuracy for non-rigid targets by incorporating geometric constraints and auxiliary scale factors. To verify the detection capability of the proposed method, we constructed a UAV Wheat Head Dataset (UWHD) and conducted extensive experiments on the UWHD, GWHD2021, and RFRB datasets. The experimental results demonstrate that ACF-YOLO outperforms other comparative methods, confirming its stable detection performance and contributing to the sustainable development of agriculture. Full article

Convolutional neural networks (CNNs) have been widely used for acne image classification due to their effectiveness in capturing local texture of skin lesions. However, the locality of convolution operations limits their ability to model long-range dependencies. Vision Transformer (ViT) methods address this issue to some extent but their high computational complexity and reliance on large-scale pre-training present challenges. Although CNN–Transformer architecture alleviates this conflict to some extent, acne images present task-specific challenges, including indistinct lesion boundaries, subtle inter-class variations, and various facial interference factors. In this paper, we propose AcneFormer, a lesion-aware and noise-robust CNN–Transformer architecture for acne image classification. We introduce three modules especially for acne tasks: a Lesion Cue Enhancement (LCE) module to highlight discriminative multi-scale spatial patterns, a Cross-Layer Feature Transmission (CLFT) module to enhance cross-layer information flow in Transformers, and a Differential Semantic Denoising (DSD) module to suppress irrelevant responses during deep feature interaction. Extensive experiments show that AcneFormer outperforms several strong baselines. Ablation and external lesion-annotated analyses further show a consistent pattern: LCE mainly improves lesion-sensitive localization and class-balanced recognition, CLFT expands valid cross-depth lesion evidence, and DSD suppresses off-lesion semantic responses. Full article

Cardiovascular disease (CVD) risk assessment in older adults requires models that are accurate, clinically interpretable, and able to retain performance in independent populations. This study developed an interpretable machine-learning framework for CVD risk stratification in individuals aged 65 years and older using routinely available clinical factors and a selected biochemical extension and then evaluated its performance in a substantially larger independent external cohort. Model development used a development cohort of 100 patients (Almaty, age ≥ 65) with leakage-free nested cross-validation and out-of-fold (OOF) probabilities. Three internally evaluated configurations were compared: a clinical logistic regression baseline (LR clinical), a biomarker-augmented logistic regression (LR selected), and a nonlinear random forest on the selected feature set (RF selected). Discrimination was assessed using ROC-AUC and PR-AUC; probabilistic accuracy using Brier score and log loss. Calibration was examined using OOF calibration curves with sigmoid calibration for selected models. Decision-analytic utility and exploratory operational thresholds were assessed using Decision Curve Analysis (DCA), yielding a three-tier scale with thresholds t__low = 0.23 and t__high = 0.40. In nested cross-validation, LR clinical achieved ROC-AUC 0.9425 ± 0.0188 and PR-AUC 0.9574 ± 0.0092 with Brier 0.1004 ± 0.0215 and log loss 0.3634 ± 0.0652; LR selected performed worse, while RF selected showed competitive discrimination. External validation on an independent cohort (n = 695) showed retained discrimination (ROC-AUC 0.8355; PR-AUC 0.9376) with acceptable probabilistic accuracy (Brier 0.1131; log loss 0.3760), and recalibration (intercept + slope) slightly improved probability metrics. Explainability analyses (odds ratios, permutation importance, SHAP) consistently identified heredity, BMI, physical activity, and diabetes as influential model-associated factors, with clinically plausible directionality. The results suggest that an interpretable model trained on a small geriatric cohort can retain meaningful predictive performance on a substantially larger external cohort, supporting the potential value of transparent risk stratification in older adults, while broader prospective and multi-center validation remains necessary before routine clinical implementation. Full article

Fungal decay fundamentally alters moisture transport in wood through complex bio-physical coupling mechanisms that remain poorly understood. Brown-rot fungi such as Coniophora puteana (Schumach.: Fr.) P. Karst. degrade wood through chelator-mediated Fenton (CMF) chemistry, producing hydroxyl radicals that depolymerise cellulose and hemicellulose before significant mass loss. This diffusion-dependent process requires elevated moisture content and leads to structural degradation. However, existing models fail to capture the interaction between boundary-driven fungal colonization, decay-induced property changes, and multi-phase multi-Fickian moisture redistribution, particularly the separate evolution of bound- and free-water phases during decay. Here, we present a transport-response bio-hygrothermal finite element model that couples boundary-driven Monod-type fungal colonization kinetics with multi-phase moisture transport (free water, bound water, vapor) in decaying wood. Although fungal biomass evolution is simulated via a reaction–diffusion equation, decay progression is not derived from biomass–substrate interaction but prescribed independently as an experimentally informed input. The model incorporates decay-modified sorption isotherms, permeability evolution, and boundary-driven biomass influx, along with associated moisture transport, into the governing equations. The model is validated against low-field nuclear magnetic resonance (LF-NMR) measurements of C. puteana decay in Scots pine over 35 days. The model successfully reproduces the experimentally observed moisture evolution: a peak free-water content of 50%–70% during weeks 1–2, followed by a progressive decline, while bound water remains remarkably constant despite advancing decay. Monte Carlo uncertainty quantification demonstrates hierarchical parameter control: bound water is governed solely by thermodynamic factors, while free water responds to interacting biological and physical processes. Time-resolved correlation analysis shows a fundamental transition from colonization-dominated (weeks 1–2) to transport-dominated (weeks 3–5) moisture control, quantitatively explaining the experimentally observed shift from accumulation to depletion. This transport-response framework for analyzing moisture behavior under externally defined decay progression establishes quantitative parameter hierarchies that may inform the development of future substrate-coupled bio-hygrothermal models. Full article

Background: Pancreatic adenocarcinoma (PAAD) is a highly malignant cancer posing severe clinical challenges. Although the dual role of pyroptosis in tumor progression is increasingly recognized, the prognostic value of its molecular heterogeneity in PAAD remains underexplored. Methods: We integrated multi-omics data and applied interpretable machine learning to construct a predictive framework centered on pyroptosis heterogeneity. Using non-negative matrix factorization (NMF) on pyroptosis-related genes (PRGs), patients were classified into distinct molecular subtypes. Evaluating 117 machine learning combinations, we employed random survival forest (RSF) to build the final model, followed by comprehensive internal and external validation. SHapley Additive exPlanations (SHAP) analysis provided global and local interpretability. Clinical potential was assessed via nomogram, drug sensitivity prediction, single-cell analysis, and immunohistochemical validation. Results: We identified two biologically distinct pyroptosis subtypes and developed a ten-gene pyroptosis subtype-associated gene signature (PSAGS). PSAGS demonstrated robust performance across training, test, and multiple external validation cohorts, outperforming most published models. Multivariate analysis confirmed its independent prognostic value, and a PSAGS-based nomogram exhibited clinical utility. PSAGS-stratified subgroups showed differential responses to immunotherapy, chemotherapy, and targeted agents. Single-cell analysis revealed cell type-specific links between PSAGS scores and pyroptosis activity, indicating that high-PSAGS malignant cells foster an immunosuppressive microenvironment through extracellular matrix (ECM)-mediated signaling. Protein-level validation confirmed upregulation of signature genes in PAAD tissues. Conclusions: This work presents a biologically reliable prognostic model for personalized PAAD management and elucidates how pyroptosis heterogeneity drives tumor progression through cellular interactions. Full article

The contemporary green transformation of the economy is a strategic imperative for businesses, especially small and medium-sized enterprises (SMEs) operating in the energy market, forcing the integration of sustainable practices in decision-making processes, including investment efficiency assessment. Classic financial tools, such as the internal rate of return (IRR) and net present value (NPV), commonly used in the SME sector, do not always adequately account for environmental, regulatory, and social risks associated with green transformation, as—particularly in the case of IRR—they rely on the assumption of stable cash flows and do not incorporate regulatory uncertainty, environmental externalities, or ESG-related risks into discounting parameters. The aim of the study was to determine the impact of nominal and real discount rates, adjusted for a synthetic measure of green transformation, on investment decisions. The research methodology combines advanced multi-criteria decision-making techniques, specifically TOPSIS and CRITIC, with sustainable finance concepts, offering an innovative approach to investment decision-making in the SME sector. The study shows that integrating environmental factors, when treated as a risk component, increases the cost of capital and reduces the net present value, while maintaining the profitability of the analysed projects. Incorporating green components into the discount rate enhances valuation appropriateness and improves investment risk management, particularly under macroeconomic uncertainty. The main contribution of the study lies in linking a synthetic green transformation indicator with dynamic discount rate adjustment within a multicriteria framework, extending existing ESG-adjusted valuation models by enabling a more structured and data-driven incorporation of environmental transition risk. Full article

To investigate the multi-factor aging mechanisms of silicone rubber used in the outer sheath of composite bushings, this study focused on HTV silicone rubber employed in the sheath layer of 1100 kV high-voltage bushings. The samples were subjected to temperature–humidity–corona coupled aging in a multi-factor aging platform. The aged samples were characterized by scanning electron microscopy, energy-dispersive spectroscopy, Fourier-transform infrared spectroscopy, hydrophobicity measurements, hardness tests, and dielectric constant measurements. The results indicate that different aging factors affect the material differently. Corona aging primarily affects the sample surface, leading to substantial methyl group detachment, surface oxidation, and a decrease in hydrophobicity, with the local static contact angle decreasing by up to 70%. In contrast, wet heat aging affects the bulk material; under high-temperature and high-humidity conditions, the internal small-molecule chains accelerate silicon-oxide crosslinking, leading to a marked increase in hardness and a relative dielectric constant that initially decreases and then increases. Considering the complex field environment, surface performance measurements are easily influenced by external factors. Therefore, hardness and relative dielectric constant are proposed as key indicators for evaluating the aging degree of silicone rubber sheaths in service. The findings provide a valuable reference for the service-life evaluation of composite bushings. Full article

Forensic age estimation is a fundamental component of biological profiling for unidentified skeletal remains, particularly in mass casualty incidents where specimens are frequently fragmented or incomplete. This review evaluates the diagnostic utility of craniofacial suture closure—specifically across four facial regions—as a non-invasive methodology for age determination in adults. By analyzing the predictable fusion patterns of ectocranial and endocranial sutures, forensic practitioners can derive approximate age ranges when postcranial indicators are absent or unreliable. Despite its utility, the reliability of suture-based estimation remains a subject of academic debate. The rate of closure is influenced by a complex interplay of environmental and biological factors, including nutritional status, hormonal influences, and mechanical loading. Historically, the method has faced criticism due to significant inter-individual variability and limited sample sizes in cadaveric studies. To improve precision and novel detail, this review explores the integration of emerging technologies such as artificial intelligence (AI) and machine learning (ML). These tools can process extensive cranial datasets to identify subtle morphological patterns that may elude human observation. While craniofacial suture analysis remains an essential resource in the forensic toolkit, its accuracy is contingent upon accounting for multi-factorial biological factors. The authors emphasize the necessity for further external validation across diverse global populations to ensure the generalizability and refinement of the technique in forensic medicine and osteology. Full article

Energy-efficient retrofit of existing residential buildings is essential for reducing heating energy demand and carbon emissions in severe cold regions. However, the absence of a structured quantitative evaluation approach often limits effective decision-making in practice. This study develops a multi-dimensional evaluation framework integrating the Fuzzy Delphi Method and Analytic Hierarchy Process (AHP) to assess and prioritize building envelope retrofit strategies. A representative non-energy-efficient residential building in Changchun, China, is selected as a case study. Based on expert consultation, a hierarchical indicator system is established, and indicator weights are determined with satisfactory consistency (CR < 0.1). The results indicate that envelope thermal performance and energy–carbon benefits are the dominant factors influencing retrofit decisions. At the parameter level, insulation thermal conductivity and external wall heat transfer coefficient are identified as the most critical variables. The findings suggest that prioritizing improvements in envelope thermal performance can effectively enhance energy-saving and carbon-reduction performance under practical constraints. The proposed framework provides a practical and transferable decision-support tool for energy-efficient retrofit planning for existing residential buildings in severe cold regions. Full article

Given the constraints inherent to island tourism resources, optimizing their allocation and utilization scientifically and efficiently has emerged as a critical challenge for both academic inquiry and policy-making. This study investigates pathways to enhance island tourism sustainability through the development of mathematical models quantifying tourism intensity, efficiency, and resource abundance, utilizing multi-source heterogeneous data on tourism resources in Hainan from 2012 to 2022. The study reveals that: (1) The spatial structure of tourism development progressed from an initial “north–south dual-core driven, fragmented in the west” pattern, through an intermediate “north–south dual-core driven, fragmented in the east” phase, and ultimately evolved into a “north–south dual-core driven, east–west isolated” configuration. (2) Spatiotemporal evolution of Hainan Island’s tourism industry is driven by a combination of policy interventions, natural endowments, transport infrastructure, economic foundations and population size. (3) Tourism economic effects exhibit marked regional heterogeneity across Hainan. Eastern regions are strongly influenced by per capita tourism income and hotel density, whereas northern areas depend more on the tertiary industry share; significant spatial spillover effects are also observed. (4) Spatial econometric modeling further indicates that influential factors do not uniformly exert positive effects on the tourism sector and its subsystems, with indirect effects exceeding direct effects by approximately 22.41 times. Although this research underscores the importance of human–environment interactions, it does not quantify the specific ecological consequences of tourism development. Future policy should integrate an ecological footprint model within a coordinated “tourism–ecology–protection” framework to balance economic and ecological goals, while also accounting for external shocks affecting the tourism economy. Full article