Search Results (4,000)

The digital transformation of financial services has increased the need for intelligent information systems capable of supporting credit risk assessment in dynamic and data-intensive environments. Traditional credit scoring approaches often face limitations when evaluating customers with limited financial histories, heterogeneous data sources, and rapidly evolving behavioral patterns. In response to these challenges, this study proposes an adaptive credit risk assessment framework that integrates machine learning, an Adaptive Champion–Challenger strategy, and MLOps practices within a unified information systems architecture. The proposed framework was evaluated using real operational data obtained from a Mexican FinTech company specializing in mobile phone financing. Three machine learning algorithms—Logistic Regression, XGBoost, and TabNet—were implemented and continuously evaluated through a rolling Champion–Challenger process supported by out-of-time validation and statistically validated model promotion criteria. Experimental results indicate that different algorithms became optimal during different evaluation periods, indicating that model effectiveness varied over time as customer behavior and portfolio characteristics evolved. While XGBoost served as the initial static baseline model, TabNet and Logistic Regression achieved superior performance during several evaluation periods, illustrating the potential benefits of adaptive model selection under changing data conditions. The proposed Adaptive Champion–Challenger Framework achieved a mean AUC of 0.817, compared with 0.798 obtained by the static baseline model. Statistical validation using the DeLong test for correlated ROC curves confirmed that the observed performance improvement was significant (p = 0.0021), providing evidence that the performance gains achieved by the adaptive strategy were unlikely to be attributable to random variation. From a Digital Transformation and Information Systems perspective, the findings suggest that maintaining predictive effectiveness in dynamic FinTech environments requires not only high-performing machine learning algorithms but also governance mechanisms that support continuous model evaluation, monitoring, traceability, and adaptive model selection. The results indicate that periodic model replacement based on statistically validated out-of-time performance can help maintain predictive effectiveness under changing data conditions while supporting model governance and operational reliability. Overall, the proposed framework provides a practical and scalable approach for implementing adaptive credit risk assessment systems that support continuous model governance, data-driven decision-making, and the management of machine learning models in alternative financing environments. Full article

Objective: The aim of this systematic review was to evaluate the in vitro effects of LED-based photobiomodulation on fibroblasts and osteoblasts, with particular focus on cellular processes involved in soft- and hard-tissue regeneration. Methods: A comprehensive electronic search was conducted on 3 April 2026 in PubMed, Scopus, Web of Science, Embase, and WorldCat according to PRISMA guidelines. The analyzed outcomes included cell viability, proliferation, migration, collagen synthesis, oxidative stress, mitochondrial activity, and selected regeneration-related processes. A total of 745 records were initially identified, and 32 studies met the inclusion criteria and were included in the qualitative synthesis. Results: The biological effects of LED-PBM depended strongly on irradiation parameters, including wavelength, fluence, irradiance, exposure time, treatment schedule, and the initial condition of the cells. Most included studies focused on fibroblast models. Red and near-infrared light showed the most consistent beneficial effects, particularly by supporting fibroblast viability, proliferation, migration, mitochondrial activity, ATP production, collagen-related responses, and oxidative stress modulation. In osteoblast-related models, LED irradiation showed potential to influence cell number, metabolic activity, maturation markers, and mineralization-related outcomes; however, the number of studies was limited. Blue light demonstrated dose-dependent effects, with higher fluences reducing fibroblast metabolic activity, proliferation, or viability. Green light improved fibroblast proliferation and migration in one model but was associated with increased cell death in osteoblast-like cells. Conclusion: LED-PBM may positively modulate cellular processes involved in soft- and hard-tissue regeneration in vitro. However, the observed effects are strongly parameter-dependent, and further standardized studies are required to define optimal irradiation protocols and validate their potential clinical relevance. Full article

The Fuzzy Cognitive Map (FCM) framework provides a useful tool for representing the complex interdependencies involved in wastewater treatment selection, particularly when social, ecological, climatic, and economic criteria are considered simultaneously. In this study, the FCM approach was applied to compare two wastewater treatment approaches in Ahvaz, Iran: constructed wetlands (CWs) as a nature-based solution and energy-based wastewater treatment plants. The developed model included 30 components and 127 causal links, and was used to examine four scenarios representing CWs, energy-based treatment, a hybrid approach, and direct wastewater discharge. The results showed that both CWs and energy-based solutions had similar effects on public health, while the hybrid scenario produced the greatest improvement. CWs had a positive effect on ecosystem restoration and showed better performance in heavy metal removal, whereas energy-based solutions had a greater negative influence on environmental conditions and climate-related components. In addition, the economic results indicated that CWs were more favorable in terms of capital cost, energy consumption, and operational cost. Sensitivity analysis using ±10% variations in causal weights showed that the main scenario-response patterns remained generally unchanged. Overall, the findings suggest that CWs and energy-based systems each have specific advantages and limitations, while the hybrid approach offers the most balanced performance across the evaluated criteria. This study demonstrates the usefulness of the FCM approach for supporting wastewater management decisions and for identifying trade-offs among treatment alternatives in sustainable water resource planning. Full article

The study addresses the design and development of a unified corporate information system for multimodal transport and logistics companies engaged in maritime and railway transportation. Analysis of the existing literature revealed the absence of a coherent methodological framework for the design of corporate information systems tailored to the specific operational characteristics of multimodal transport and logistics enterprises. To bridge this gap, a design methodology for corporate information system databases is proposed, intended for subsequent deployment in companies operating multimodal supply chains. The development of the unified corporate information system was guided by the principle of “total costs,” which requires that the decision-maker—when selecting transport modes, methods of transportation, carriers, routing, and auxiliary intermediaries (insurer, stevedore, bank, and customs broker)—address the problem as an integrated whole rather than optimizing individual components in isolation. The study encompasses information modeling of the business processes of multimodal transport and logistics companies, construction of an optimal model of the transport process for maritime and railway transportation using integrated computer automated manufacturing definition (IDEF) and structured analysis and design technique (SADT) modeling, and the design of a multilevel unified database structure for the coordination of different transport modes. A decision-making and support system has been developed for managing the operational activities of a multimodal transport and logistics company engaged in maritime and railway transportation. The proposed unified corporate information system enables the replacement of task resolution by local optimization criteria applied separately to each transport mode—such as freight cost and delivery time—with a single global optimization criterion for the multimodal supply chain. Full article

The concept of ranked sparsity, originally introduced in the context of penalized regression, arises in modeling applications when an expected disparity exists in the quality of information between different feature sets. Its presence can cause traditional and modern model selection methods to fail because such procedures commonly presume “covariate equipoise”—that each potential parameter is equally worthy of entering into the final model. However, this presumption does not always hold, especially in the presence of derived variables or with highly disparate feature sets (i.e., multi-modal data). For instance, when all possible interactions are considered as candidate predictors, the sheer number of them grossly inflates the number of false discoveries, resulting in unnecessarily complex and difficult-to-interpret models with many (truly spurious) interactions. In this work, we motivate a ranked sparsity extension to the Bayesian Information Criterion (RBIC) that requires a stronger level of evidence in order to allow certain variables (e.g., interactions vs main effects and genetic vs clinical covariates) into a model. We compare the performance of RBIC relative to competing methods for selecting polynomials and interactions in a simulation study and in two applications, showing that stepwise selection guided by RBIC produces better-predicting, more transparent models (with fewer false interactions) compared to existing alternatives. Full article

Background: Delirium management remains inconsistently implemented in intensive care units (ICUs), and existing studies provide limited insight into how assessment capability, organisational opportunity, and emotional motivation interact in Chinese tertiary care settings. Aim: To explore ICU nurses’ perceived barriers and facilitators to delirium assessment and management using the Capability, Opportunity and Motivation model of Behaviour (COM-B). Methods: A qualitative descriptive exploratory study was conducted with 14 registered ICU nurses from two tertiary hospitals in Chongqing, China, between December 2025 and February 2026. Purposive sampling captured variation in clinical experience, educational background, and ICU type. The first author conducted face-to-face interviews in Mandarin Chinese and completed the initial coding. COM-B informed both the interview guide and the deductive organisation of the analysis, while six themes and 33 lowest-level factors were developed within the three domains. During revision, the fixed analytic structure was audited against all 14 Mandarin transcripts, and the second author independently applied the fixed codebook to five complete transcripts selected for maximum variation; disagreements were resolved by returning to the original Mandarin context and predefined coding boundaries. Results: Eighteen barriers and fifteen facilitators were identified across six themes within the COM-B domains. Capability was strengthened by clinical experience, senior nurses’ intuitive pattern recognition, and conversational assessment, but constrained by difficulty recognising hypoactive delirium, uncertainty when assessing older adults, inconsistent use of CAM-ICU, and insufficient targeted training. Opportunity was enhanced by staffing, environmental control, peer support, family engagement, and physician collaboration, but restricted by fragmented visibility, sensory overstimulation, workload, restrictive visiting, and delayed medical responses. Motivation reflected the tension between professional duty and accountability for safety on one hand and emotional burden, verbal abuse, and violence on the other. Cross-domain accounts showed that opportunity constraints could amplify anxiety and safety-first motivation, narrowing feasible behaviour toward containment, whereas social support could sustain relational care. Conclusions: Delirium management was shaped by interacting behavioural, organisational, and emotional conditions. Strengthening assessment capability alone is unlikely to produce consistent practice without staffing, workflow, environmental, and interprofessional support. Mapping the findings to Behaviour Change Wheel intervention functions highlighted training, environmental restructuring, enablement, and modelling as priority candidate functions for prospective evaluation; education alone was unlikely to be sufficient. Reporting Method: This study was reported in accordance with the Consolidated Criteria for Reporting Qualitative Research checklist. Patient or Public Contribution: No patient or public contribution was involved in the design, conduct or reporting of this study. Participants were registered intensive care unit nurses. Full article

Objective: This study investigates the relationship between social media use and consumer purchase intentions or out-of-home food choices in the Polish gastronomic market, focusing specifically on variations across generations. Methodology: Data were collected between May and July 2024 through a diagnostic survey using the Computer-Assisted Web Interviewing (CAWI) technique (N = 1099). Respondents were recruited via a non-probability quota sampling approach based on strict demographic inclusion criteria. To test the research hypotheses regarding generational variations in market behaviour, a multivariate discriminant function analysis was performed using Statistica 13.1 PL. A preliminary pilot study (N = 30) confirmed the initial questionnaire readability and overall consistency (Cronbach’s alpha = 0.87), while subscales were treated independently during the main analysis. Results: Multivariate models were found to be highly significant, revealing clear differences between age groups. The youngest cohort (aged 18–35) relies heavily on Instagram and TikTok, showing distinct patterns regarding visual triggers such as “instagrammable” aesthetics, digital validation, and menu uniqueness alongside weight loss claims. Conversely, seniors (aged 61 and older) lean unexpectedly towards X. This older segment displays pragmatic, utility-driven motives, searching for detailed textual data about ingredients and the health-promoting properties of food. General product quality and calorie control were identified as universal factors that do not vary by generation. Conclusions and Managerial Implications: The digital transformation of the Polish restaurant industry does not follow a single path. Social media is closely linked to modern customer journeys, with physical dining spots frequently serving as spaces for socialisation. Consequently, restaurant operators should move away from mass communication and adopt a selective omnichannel strategy, where message formats shift from visual appeal to factual, nutrition-oriented text that is tailored to the digital literacy and dietary needs of each generation. Full article

The extraction of spatiotemporal features from video sequences allows for the recognition of actions and the analysis of behaviors in video, making it a key challenge in automated video processing. The literature shows widespread use of deep learning approaches, specifically convolutional neural networks (CNNs). In this context, researchers face the challenge of identifying the advantages, disadvantages, and emerging trends across different architectures, evaluation metrics, and even dataset selection. The objective of this study is to identify the most common CNN architectures, evaluation metrics, datasets, and trends in spatiotemporal feature extraction for video analysis. The selection of articles used the PRISMA methodology and the Joanna Briggs Institute (JBI) methodological framework. From the databases Scopus, Web of Science and the MDPI platform, and based on the inclusion/exclusion criteria, 31 articles that met the criteria were analyzed and synthesized. The search was conducted primarily using the keywords “Convolutional Neural Network,” “video processing,” and “feature extraction,” limiting the selected works to those published between 2020 and the end of 2025. The results mainly show the use of four neural network architectures: 2D CNNs, 3D CNNs, hybrid models (e.g., CNN–RNN, CNN–Transformer, and multi-stream models), and, to a lesser extent, lightweight architectures. Commonly used datasets were identified (e.g., UCF101 and HMDB51). Additionally, standardized evaluation metrics were identified, ranging from accuracy and F1-score to performance measures specific to each case study. The challenges identified center on the heterogeneity of the study datasets, the lack of standardized evaluation metrics, and maintaining a balance between accuracy and computational resource consumption. On the other hand, strong emerging trends toward the use of hybrid models and those integrating transformers have been identified. This systematic review emphasizes the need for clear and robust guidelines that allow for the appropriate selection of CNN architecture, test datasets, and evaluation metrics in applications for extracting spatiotemporal features from video sequences, as well as identifying trends and future lines of research. Full article

Currently, insulation faults in the DC system of photovoltaic (PV) power stations are handled by a full shutdown strategy of inverters, and fault branch localization relies on manual inspection, resulting in low efficiency and poor accuracy, leading to prolonged unplanned outages and substantial power generation losses. This paper proposes an integrated solution combining high-precision insulation monitoring and intelligent fault line selection, which ensures the reliability of line selection criteria through improved measurement accuracy and achieves automatic fault isolation via optimized line selection strategies. The paper analyzes the mathematical essence of the ill-conditioned measurement equations of the traditional bridge method under severe single-pole grounding faults, establishes a dual-channel heteroscedastic noise model, and utilizes the inherent physical constraint that the sum of the positive and negative pole-to-ground voltages always equals the bus voltage to transform the ill-posed inverse problem into an equality-constrained optimal estimation problem, deriving an analytical solution in the sense of constrained least squares. A collaborative monitoring strategy of “balanced bridge monitoring first, unbalanced bridge precision measurement afterward” is proposed. An automatic fault line selection and isolation algorithm based on sequential branch switching is designed, which leverages the operational characteristic that PV systems allow short-term branch interruption, enabling automatic identification and isolation of faulty branches and automatic restoration of non-faulty branches without installing any leakage current sensors. Experimental results show that under severe fault conditions with a single-pole insulation resistance as low as 22 kΩ, the proposed method limits the error to within 5%; the proposed line selection strategy can complete identification and isolation of all faulty branches within at most two rounds of switching. Full article

High-entropy alloys (HEAs) are a unique metallic alloy that was initially recognized for the possibility of stabilizing simple solid-solution phases through high configurational entropy. Research over the past two decades, however, has shown that their behavior is far more complex. Phase formation, microstructural evolution, and surface performance arise from the combined influence of composition, atomic interactions, processing history, and the surrounding environment. This paper reviews the connections between these aspects moving from the bulk alloy to the surface. The thermodynamic and empirical criteria utilized for prediction of phase formation and reasons behind ignoring the factors such as ordering, segregation, metastability, and processing defects are described. Further, the influence of casting, rapid solidification, coating deposition, and thin-film processing on the microstructure that will interact with catalytic or corrosive environment is reviewed. Electrocatalysis and corrosion resistance are considered as two strongly coupled surface phenomena rather than separate fields of application. Quantitative comparison of exemplary high-entropy alloy systems shows the influence of the alloying approach and surface development on the catalytic properties, surface reconstruction, selective dissolution, passive film formation, and localized corrosion. The potential of CALPHAD modeling, density functional theory, machine learning, and multi-objective optimization for a better alloy selection in the field of high-entropy alloys is reviewed as well. We identified that the success of HEA design is not only in choosing the right composition but rather in controlling the phases, defects, interfaces, and surface of the HEA. Full article

Although deep learning methods have shown high performance in chest X-ray classification, high accuracy alone does not guarantee reliable reasoning. A model may still exhibit pathological behavior, such as unstable evidence usage under harmless input changes, inconsistent reasoning across augmented views, excessive dependence on surrounding frame information, and appearance bias dominance, where prediction relies too heavily on intensity while neglecting texture and shape. In this paper, we propose a bio-inspired pathology-aware, factor-aware framework for explainable and reliable chest X-ray classification, inspired by biological vision principles such as figure–ground separation, selective attention, and balanced use of complementary visual cues. During training, the method regularizes appearance bias dominance through evidence-guided counterfactual perturbations that mimic cue-suppression analysis in biological perception, thereby revealing and penalizing excessive factor dependence. During testing, it evaluates model behavior using four criteria: reasoning stability, augmentation inconsistency, appearance bias dominance, and framing sensitivity. This combination enables the framework to go beyond conventional inference-time explanation by both correcting pathological behavior during training and exposing it during evaluation. From a biomimetic perspective, the framework encourages the model to separate relevant foreground anatomy from surrounding background and to avoid over-reliance on a single dominant cue. The proposed approach improves interpretability and reliability without modifying the backbone architecture or increasing model size or inference-time cost. The proposed training process improved the F1-score of DenseNet-121 from 0.899 to 0.931, while also producing more stable and balanced reasoning. Full article

Environmental, Social, and Governance (ESG) disclosure plays an increasingly important role in evaluating corporate sustainability performance. However, the agricultural chemicals industry faces unique environmental and social challenges, while existing ESG assessment frameworks remain largely generic and fail to capture industry-specific disclosure requirements. To address this gap, this study develops a multi-level ESG disclosure quality evaluation framework for the agricultural chemicals industry based on the GRI 2021 Standards, China’s regulatory requirements, and sector-specific production characteristics. The framework was developed through targeted qualitative content analysis and text coding of ESG disclosures from ten listed agricultural chemical companies selected from the complete eligible population of 17 Chinese A-share agricultural chemicals enterprises that met the study’s predefined inclusion criteria and had Huazheng ESG ratings. This process resulted in a hierarchical structure comprising three dimensions, 14 first-level indicators, 61 second-level indicators, and 350 third-level observation indicators, with particular emphasis on biodiversity conservation, farmer support, and corporate governance. The Analytic Hierarchy Process (AHP) was then applied to determine the weights of the first-level and second-level indicators through expert evaluation, while a three-point scoring system (0–2) was established for the third-level indicators to construct the industry-specific ESG disclosure quality evaluation model. The framework was subsequently evaluated using an independent sample of the remaining seven listed agricultural chemical companies. A benchmarking comparison with Huazheng ESG ratings showed broad alignment in overall patterns while also revealing important company-level differences and disclosure-quality gaps not readily captured by the general ESG ratings. Furthermore, multi-level ±20% weight perturbation analyses demonstrated the framework’s stability, discriminative ability, industry suitability, and computational robustness. The proposed framework provides a practical and transparent tool for assessing ESG disclosure quality in the agricultural chemicals industry and offers a methodological foundation for developing sector-specific ESG disclosure evaluation frameworks in other high-impact industries. Full article

Background: Arrhythmia recurrence after cryoballoon ablation remains a clinically relevant problem in atrial fibrillation (AF). However, factors associated with documented recurrence in low-risk patients without major comorbidities are not well defined. This study aimed to evaluate factors associated with documented arrhythmia recurrence after cryoballoon ablation in a highly selected low-risk AF population. Methods: This retrospective, single-center study included 153 eligible patients selected from an institutional cryoablation database after application of predefined exclusion criteria. Only patients with a CHA₂DS₂-VA (congestive heart failure, hypertension, age ≥ 75 years, diabetes mellitus, stroke/transient ischemic attack/thromboembolism, vascular disease, and age 65–74 years) score ≤ 1 and without major comorbidities, including diabetes mellitus, hypertension, coronary artery disease, chronic kidney disease, cerebrovascular disease, and heart failure, were included. Recurrence was defined as electrocardiographically documented AF or atrial tachyarrhythmia after the 3-month blanking period over 1 year of follow-up; Holter-detected episodes were required to last ≥30 s. The primary multivariable model included sex, age, AF type, and absolute left atrial diameter; a sensitivity model replaced absolute left atrial diameter with left atrial diameter indexed to body surface area (BSA). Results: The study population included 70 women (45.8%) and 83 men (54.2%). Paroxysmal AF was present in 125 patients (81.7%), whereas 28 patients (18.3%) had persistent AF. Documented arrhythmia recurrence occurred in 40 patients (26.1%). Female sex was more frequent in the recurrence group than in the no-recurrence group (62.5% vs. 39.8%, p = 0.013). The complete-case primary multivariable model included 126 patients with 36 recurrence events. Female sex was associated with documented arrhythmia recurrence (odds ratio [OR] 2.69, 95% confidence interval [CI] 1.14–6.37; p = 0.024). The estimate for left atrial diameter was directionally positive but statistically uncertain (OR 1.086 per mm, 95% CI 0.989–1.192; p = 0.085). In the BSA-indexed sensitivity model (n = 123), the female-sex estimate was attenuated and statistically uncertain (OR 2.16, 95% CI 0.90–5.18; p = 0.084), while left atrial diameter/BSA was also statistically uncertain (OR 1.145 per mm/m², 95% CI 0.977–1.343; p = 0.095). Conclusions: In this selected low-risk cohort undergoing second-generation cryoballoon ablation, female sex was associated with clinically detected, electrocardiographically documented arrhythmia recurrence in the primary model under an intermittent rhythm-surveillance strategy based on scheduled 12-lead electrocardiograms (ECGs) and symptom-driven evaluations. However, the estimate was attenuated and statistically uncertain after indexing left atrial diameter to BSA. Given the retrospective, single-center design, non-systematic rhythm monitoring, and sensitivity of the sex estimate to body-size adjustment, these findings should be interpreted as hypothesis-generating. Full article

Photovoltaic technology selection in Gulf Cooperation Council (GCC) desert environments is affected by coupled dust, thermal, ultraviolet (UV), humidity, and salinity stresses, which are not fully represented by static weighting and additive multi-criteria decision-making models. This study develops a GCC evidence-calibrated nonlinear decision-support framework that integrates published literature-derived GCC/desert-stress calibration, adaptive hybrid entropy–desert weighting, and bipolar fuzzy Einstein aggregation. The framework is applied to compare passivated emitter and rear cell (PERC), tunnel oxide passivated contact (TOPCon), and heterojunction technology (HJT) photovoltaic technologies using calibrated evidence from Qatar, the United Arab Emirates, Saudi Arabia, and Oman. The results show that dust tolerance receives the highest final hybrid weight (0.258), followed by thermal resistance (0.228), UV resistance (0.207), efficiency (0.173), and cost effectiveness (0.134). The nonlinear Einstein aggregation ranks HJT first (0.889), followed by TOPCon (0.861) and PERC (0.742). Benchmark comparison with TOPSIS, VIKOR, and PROMETHEE II shows high rank agreement, while Monte Carlo perturbation analysis indicates that HJT preserves the first rank in 93% of perturbation runs. The proposed framework links PV technology selection with published GCC desert-stress evidence and provides a reproducible basis for technology prioritization in harsh solar energy deployment environments. A stress-to-decision translation table is also provided to clarify how desert degradation mechanisms are converted into decision criteria and reusable selection guidance. Full article

Background: Composite anthropometric and cardiometabolic indices have emerged as practical tools for identifying metabolic syndrome (MetS), but their comparative classification performance against ALAD-defined MetS has not been comprehensively evaluated in Panamanian adults. This study compared the classification performance of selected anthropometric, body composition, and cardiometabolic indices for identifying metabolic syndrome defined according to the Latin American Diabetes Association (Asociación Latinoamericana de Diabetes [ALAD]) criteria in non-diabetic Panamanian adults. Methods: This cross-sectional study included 262 adults without diabetes (110 men and 152 women). Clinical and laboratory measurements were used to calculate the body mass index (BMI), waist-to-height ratio (WHtR), visceral adiposity index (VAI), lipid accumulation product (LAP), triglyceride-to-high-density lipoprotein cholesterol ratio (TG/HDL-C), atherogenic index of plasma (AIP), cardiometabolic index (CMI), and Homeostasis Model Assessment of Insulin Resistance (HOMA-IR). Classification performance was evaluated using receiver operating characteristic (ROC) curve analysis, optimal cut-off values, and DeLong’s test. Age-adjusted logistic regression assessed associations between each index and ALAD-defined MetS. Results: Composite indices integrating anthropometric and metabolic parameters showed high classification performance. CMI had the highest observed AUC in men (AUC = 0.933; 95% CI: 0.887–0.979), whereas LAP had the highest observed AUC in women (AUC = 0.854; 95% CI: 0.778–0.931). Conclusions: CMI and LAP showed high classification performance for identifying ALAD-defined MetS in Panamanian adults without diabetes, with the highest observed AUCs in men and women, respectively. These simple, inexpensive indices may have potential utility as screening tools for identifying ALAD-defined MetS in primary healthcare settings. The proposed cut-off values are exploratory and require external validation in independent Panamanian and Latin American populations before clinical implementation. Full article