Search Results (9,598)

Architectural design stands as a highly knowledge-intensive field, with designers serving as the linchpin for its premium development. China’s construction industry is now navigating a transitional phase of slower growth, where a misalignment in designer capabilities significantly obstructs the nation’s shift from being a mere “construction giant” to becoming a true “construction powerhouse”. Based on the spatial mismatch model and Geodetector, this study empirically analyzes the mismatch relationship among designers and its influencing factors using panel data from 31 provinces in China from 2013 to 2023, and proposes strategies for cultivating architectural design talents. Findings reveal that China’s architectural designers exhibit spatial supply imbalance, and complex trends in designer allocation-simultaneous growth and decline coexist. China exhibits diverse types of architect mismatch: 22.58% of regions are in a state of Positive Mismatch, and 12.90% experience Negative Mismatch. In over one-third of regions, the architectural design talent market can no longer self-correct architect mismatch through market mechanisms, urgently requiring collaborative intervention policies from governments, design associations, and enterprises to address architect supply–demand governance. For a smooth transition during the transformation and upgrading of the construction and design industries, the architectural design talent market should accommodate frictional designer mismatch. The contribution of designer mismatch varies significantly, with factors such as innovation, industrial structure, and fiscal self-sufficiency exerting more direct influence, while other factors play indirect roles through dual-factor enhancement effects and nonlinear enhancement effects. The insights from the analysis results and conclusions for future designer cultivation include fostering an interdisciplinary teaching model for designers through university–enterprise collaboration, enhancing education in AI and intelligent construction literacy, and establishing an intelligent service platform for designer supply–demand matching to promptly build a new differentiated and precise designer supply system. Full article

Biometric authentication, such as facial recognition and fingerprint scanning, is now standard on mobile devices, offering secure and convenient access. However, the processing of biometric data is tightly regulated under the European Union’s General Data Protection Regulation (GDPR), where such data qualifies as “special category” personal data when used for uniquely identifying individuals. Compliance requires meeting strict conditions, including explicit consent and data protection by design. Passkeys, the modern name for FIDO2-based authentication credentials developed by the FIDO Alliance, enable passwordless login using public key cryptography. Its “match-on-device” architecture stores biometric data locally in secure hardware (e.g., Android’s Trusted Execution Environment, Apple’s Secure Enclave), potentially reducing the regulatory obligations associated with cloud-based biometric processing. This paper examines how Passkeys are implemented on Android and iOS platforms and their differences in architecture, API access, and hardware design, and how those differences affect compliance with the GDPR. Through a comparative analysis, we evaluate the extent to which each platform supports local processing, data minimization, and user control—key principles under GDPR. We find that while both platforms implement strong local protections, differences in developer access, trust models, and biometric isolation can influence the effectiveness and regulatory exposure of Passkeys deployment. These differences have direct implications for privacy risk, legal compliance, and implementation choices by app developers and service providers. Our findings highlight the need for platform-aware design and regulatory interpretation in the deployment of biometric authentication technologies. This work can help inform stakeholders, policymakers, and legal experts in drafting robust privacy and ethical policies—not only in the realm of biometrics but across AI technologies more broadly. By understanding platform-level implications, future frameworks can better align technical design with regulatory compliance and ethical standards. Full article

Varroa mites are a major global threat to honeybee colonies. Combining digital twins with scenario-generating models can be an enabler of precision apiculture, allowing for monitoring Varroa spread, generating treatment scenarios under varying conditions, and running remote interventions. This paper presents the conceptual design of this system for large-scale Varroa management in honeybee apiaries, with initial validation conducted through simulations and feasibility analysis. The design followed a design research framework. The proposed system integrates a wireless sensor network for continuous hive sensing, image capture, and remote actuation of treatment. It employs generative time-series models to forecast colony dynamics and a statistical network model to represent inter-colony spread; together, they support spread scenario prediction and what-if evaluations of treatments. The system evolves through continuous updates from field data, improving the accuracy of spread and treatment models over time. As part of our design research, an early feasibility assessment was carried out through the generation of synthetic data for spread model pretraining. In addition, a node-level energy budget for sensing, communication, and in-hive treatment was developed and matched with battery capacity and life calculations. Overall, this work outlines a path toward real-time, data-driven Varroa management across apiary networks, from regional to cross-border scales. Full article

Accurate detection of fetal distress from cardiotocography (CTG) is clinically critical but remains subjective and error-prone. In this research, we present a leakage-safe Calibrated Global Logit Fusion (CGLF) framework that couples TabNet’s sparse, attention-based feature selection with XGBoost’s gradient-boosted rules and fuses their class probabilities through global logit blending followed by per-class vector temperature calibration. Class imbalance is addressed with SMOTE–Tomek for TabNet and one XGBoost stream (XGB–A), and class-weighted training for a second stream (XGB–B). To prevent information leakage, all preprocessing, resampling, and weighting are fitted only on the training split within each outer fold. Out-of-fold (OOF) predictions from the outer-train split are then used to optimize blend weights and fit calibration parameters, which are subsequently applied once to the corresponding held-out outer-test fold. Our calibration-guided logit fusion (CGLF) matches top-tier discrimination on the public Fetal Health dataset while producing more reliable probability estimates than strong standalone baselines. Under nested cross-validation, CGLF delivers comparable AUROC and overall accuracy to the best tree-based model, with visibly improved calibration and slightly lower balanced accuracy in some splits. We also provide interpretability and overfitting checks via TabNet sparsity, feature stability analysis, and sufficiency (k95) curves. Finally, threshold tuning under a balanced-accuracy floor preserves sensitivity to pathological cases, aligning operating points with risk-aware obstetric decision support. Overall, CGLF is a calibration-centric, leakage-controlled CTG pipeline that is interpretable and suited to threshold-based clinical deployment. Full article

Sport performance results from the interaction of tactical, technical, physiological and psychological factors, but psychological aspects are often minimized or analyzed in a decontextualized manner. This exploratory pilot study aimed to contribute to the development of a diagnostic framework that links individual behaviors during football attack–defense transition moments (ADT) with psychological attributes. Twenty male U14 players were assessed across five official matches regarding their ADT performance indicators. The Athletic Coping Skills Inventory (ACSI-28) and the Resilience Scale (RS) were applied during the competition. Statistical analyses included correlation tests and Bayesian analysis. Players showed a significant tendency to sustain ball recovery behaviors after possession loss (p = 0.004). Psychological resilience and athletic coping skills varied substantially between individuals without positional differences, as well as RS scores were significantly below the high-resilience threshold (147; p = 0.013). A moderate positive correlation emerged between RS Factor 1 and the ACSI-28 subscale “Coping with Adversity” (r = 0.574, p = 0.008). Posterior distributions provide exploratory signals suggesting possible positive associations for two psychological constructs considering ADT individual behaviors: “Concentration” in relation to the maintenance of recovery actions (Mode = 0.439; 95% CI [0.030, 0.721]) and “Goal Setting” in relation to the rapid initiation of recovery actions (Mode = 0.465; 95% CI [0.059, 0.734]). Nevertheless, Bayes Factors favored the null model overall, indicating that these signals are weak and require replication. By contrast, most psychological constructs, including resilience, showed no reliable evidence of correlation with recovery-related actions. The findings highlight the need to further research the integration of psychological assessment into football performance diagnostics, while also indicating that psychological factors alone are insufficient to fully explain youth players’ individual ADT behaviors. Full article

A good merger and acquisition (M&A) cannot be achieved without a good matching that not only ensures high satisfaction for both bidders and targets but also operates as a two-sided process based on their mutual evaluations. Previous studies mostly focus on estimating the potential gains, and even those addressing M&A matching or the selection of merger targets and partners overlook the herd behavior of decision makers in the mutual evaluation. Nevertheless, decision makers often adjust their opinions by consulting others’ opinions, especially those they trust, and behave bounded rationality. Based on these, we propose a new approach for the two-side M&A matching from a perspective based on the mutual performance evaluation considering herd behavior. First, based on the concept of the cross efficiency in data envelopment analysis field, we propose a mutual evaluation method considering herd behavior of bidders and targets. Then, we measure the bidders’ and targets’ satisfaction with each other based on the prospect theory. Next, to seek the optimal M&A matching strategy, we build a two-side matching model with two objective functions that maximize the bidders’ and targets’ satisfaction with each other simultaneously. Finally, we use the data of 51 banks to illustrate our method. Full article

Background and Objectives: Acute pulmonary thromboembolism (PTE) is one of the leading causes of cardiovascular mortality. Recent insights into PTE pathophysiology emphasize the complex interplay of multiple mechanisms, particularly the roles of thrombosis and inflammation. Methods: This retrospective, single-center observational study included 138 participants: 69 adult patients diagnosed with PTE via computed tomography pulmonary angiography (CTPA) and 69 matched healthy controls. Upon admission, a standard 12-lead electrocardiogram (ECG) was performed, and Daniel’s score was calculated. Peripheral blood samples were collected to assess inflammatory biomarkers and hemogram-derived ratios (SII, NLR, dNLR, NPR, PLR, LMR). CTPA scans were analyzed not only for diagnostic purposes and PTE localization but also for inflammatory changes. PTE severity was classified according to the 2019 ESC guidelines. Results: Patients with PTE had significantly higher Daniel’s ECG scores, initial values of inflammatory biomarkers (WBC, neutrophils, IL-6, CRP) and hemogram-derived ratios (SII, NLR, dNLR, NPR) compared to controls. In multivariate analysis, older age (OR = 1.05; p = 0.038), higher Daniel’s ECG score (OR = 1.24; p < 0.001), and higher dNLR (OR = 1.40; p = 0.001) were found as an independent predictors of PTE severity. Ground-glass opacity (GGO) was the most common parenchymal and pleural inflammatory finding relating to CTPA (48.4%), but these findings did not show significant predictive value for PTE severity. Conclusions: Daniel’s ECG score and dNLR, both readily available and cost-effective biomarkers demonstrated independent predictive value for assessing PTE severity. Full article

UAV accident investigation is essential for safeguarding the fast-growing low-altitude airspace. While near-daily incidents are reported, they were rarely analyzed in depth as current inquiries remain expert-dependent and time-consuming. Because most jurisdictions mandate formal reporting only for serious injury or substantial property damage, a large proportion of minor occurrences receive no systematic investigation, resulting in persistent data gaps and hindering proactive risk management. This study explores the potential of using large language models (LLMs) to expedite UAV accident investigations by extracting human-factor insights from unstructured narrative incident reports. Despite their promise, the off-the-shelf LLMs still struggle with domain-specific reasoning in the UAV context. To address this, we developed a human factors analysis and classification system (HFACS)-guided analytical framework, which blends structured prompting with lightweight post-processing. This framework systematically guides the model through a two-stage procedure to infer operators’ unsafe acts, their latent preconditions, and the associated organizational influences and regulatory risk factors. A HFACS-labelled UAV accident corpus comprising 200 abnormal event reports with 3600 coded instances has been compiled to support evaluation. Across seven LLMs and 18 HFACS categories, macro-F1 ranged 0.58–0.76; our best configuration achieved macro-F1 0.76 (precision 0.71, recall 0.82), with representative category accuracies > 93%. Comparative assessments indicate that the prompted LLM can match, and in certain tasks surpass, human experts. The findings highlight the promise of automated human factor analysis for conducting rapid and systematic UAV accident investigations. Full article

The rapid growth of digital communication has intensified spam-related threats, including phishing and malware, which employ advanced evasion tactics. Traditional filtering methods struggle to keep pace, driving the need for sophisticated machine learning (ML) solutions. The effectiveness of ML models hinges on selecting high-quality input features, especially in high-dimensional datasets where irrelevant or redundant attributes impair performance and computational efficiency. Guided by principles of symmetry to achieve an optimal balance between model accuracy, complexity, and interpretability, this study proposes an Enhanced Hybrid Quantum-Inspired Firefly and Artificial Bee Colony (EHQ-FABC) algorithm for feature selection in spam detection. EHQ-FABC leverages the Firefly Algorithm’s local exploitation and the Artificial Bee Colony’s global exploration, augmented with quantum-inspired principles to maintain search space diversity and a symmetrical balance between exploration and exploitation. It eliminates redundant attributes while preserving predictive power. For interpretability, Shapley Additive Explanations (SHAPs) are employed to ensure symmetry in explanation, meaning features with equal contributions are assigned equal importance, providing a fair and consistent interpretation of the model’s decisions. Evaluated on the ISCX-URL2016 dataset, EHQ-FABC reduces features by over 76%, retaining only 17 of 72 features, while matching or outperforming filter, wrapper, embedded, and metaheuristic methods. Tested across ML classifiers like CatBoost, XGBoost, Random Forest, Extra Trees, Decision Tree, K-Nearest Neighbors, Logistic Regression, and Multi-Layer Perceptron, EHQ-FABC achieves a peak accuracy of 99.97% with CatBoost and robust results across tree ensembles, neural, and linear models. SHAP analysis highlights features like domain_token_count and NumberOfDotsinURL as key for spam detection, offering actionable insights for practitioners. EHQ-FABC provides a reliable, transparent, and efficient symmetry-aware solution, advancing both accuracy and explainability in spam detection. Full article

Background and Objectives: This study is aimed at evaluating periodontal health in patients with psoriasis and investigating its impact on dermatology-specific and oral health-related quality of life. Materials and Methods: A total of 226 individuals were enrolled, including 113 patients with clinically diagnosed psoriasis and 113 age- and gender-matched healthy controls. The periodontal parameters recorded included plaque index (PI), gingival index (GI), probing depth (PD), and clinical attachment loss (CAL). Oral health-related quality of life was assessed using the Oral Health Impact Profile-14 (OHIP-14), while dermatology-specific quality of life was evaluated with the Psoriasis Quality of Life Questionnaire (PQLQ). Psoriasis severity was measured by the Psoriasis Area and Severity Index (PASI). Results: Patients with psoriasis demonstrated significantly poorer periodontal parameters compared to controls, with higher PI (p = 0.006), PD (p = 0.001), and CAL (p = 0.041), as well as a lower number of teeth (p = 0.027). No significant differences in GI were observed (p = 0.331). Subdomain analysis of OHIP-14 indicated significantly greater functional limitation in the psoriasis group (p = 0.001), although no differences were detected in other domains. Positive and significant correlations were found among all the OHIP-14 subscales in both groups, and PQLQ scores were strongly correlated with OHIP-14 outcomes in the psoriasis group (p < 0.05). PASI scores tended to be higher among patients with periodontitis than those with gingivitis or periodontal health, but this difference did not reach statistical significance (p = 0.257). Conclusions: Psoriasis patients exhibited poorer periodontal status and reduced oral health-related quality of life compared to healthy individuals. However, differences in oral hygiene habits may also have contributed to these findings. Our findings suggest an association between psoriasis and impaired periodontal health, but due to the cross-sectional design, a causal relationship cannot be established. Full article

Background/Objectives: Patients with single metastases from colorectal cancer constitute a subgroup with an excellent 5-year OS of 55–70% and with a real chance for cure. In this situation, local margin recurrence in the lung may impair the prognosis and thus is the main outcome target of surgery. Methods: A retrospective multicenter analysis of patients with single metastases from colorectal cancers was performed. Four German Thoracic Surgery units contributed data from their prospective metastasectomy databases. Statistical analysis was focused on tumor recurrence and risk factors for local margin recurrence. Results: 166 patients from four centers could be further analyzed. For later comparison, 93 (56%) anatomic resections and 73 (44%) non-anatomic resections were pooled. Tumor recurrence was detected: at any site 87/161 (54%), within the lung 62/161 (38.5%) at intrapulmonary margins 25/145 (17.2%) and in intrathoracic lymph nodes 14/138 (10.1%). Intrapulmonary local margin recurrence was more often found in non-anatomic (25.4%) versus anatomic (11.6%) resections (p = 0.052). After propensity score matching (PSM), local margin recurrence was significantly more frequent after non-anatomic resection of intermediate and peripherally located metastases (p = 0.042). Furthermore, local margin recurrence was associated with small safety margins (p < 0.001), small number of lymph nodes removed (p < 0.001) and with intrathoracic lymph node recurrence (p = 0.001). The 5- and 10-year OS of the whole group was 70% and 47% with a median survival of 9.0 years. The 5- and 10-year RFS of the whole group was 59% and 43% with a median of 7.3 years. Conclusions: This study demonstrates that anatomical resection of single CRC lung metastases is superior to non-anatomic resection with respect to local radicality and local intrapulmonary margin recurrence, but there was no difference in OS and RFS. Full article

Background and Objectives: Minimally invasive surgery is considered the standard of care for early-stage endometrial cancer. However, the oncological safety of intrauterine manipulator (IUM) use during laparoscopic hysterectomy remains controversial. The aim of this study was to evaluate the impact of intrauterine manipulator use during laparoscopic hysterectomy on oncological outcomes in patients with clinically early-stage endometrial cancer. Materials and Methods: In this retrospective cohort study, 612 patients with FIGO 2009 stage I–III endometrial cancer who underwent staging surgery at a tertiary center between January 2010 and May 2025 were included. Clinical and pathological characteristics were compared between laparoscopy (n = 168) and laparotomy (n = 444). To reduce selection bias, propensity score matching (PSM) was performed based on age, histological subtype, and FIGO stage. Kaplan–Meier survival analysis and Cox regression modeling were utilized to evaluate disease-free survival (DFS) and overall survival (OS). Results: After matching, groups were balanced except for higher rates of para-aortic lymphadenectomy and adjuvant therapy in the laparotomy group. IUM use was not associated with increased LVSI or positive peritoneal cytology. Recurrence was more frequent after laparoscopy (10.1% vs. 6.0%, p = 0.028), with inferior 5-year DFS (87.6% vs. 97.4%, HR 5.60, p = 0.0006), while OS was similar (82.0% vs. 87.6%, p = 0.842). In multivariate Cox analysis, independent predictors of worse DFS were non-endometrioid histology (HR 3.57), FIGO stage III (HR 3.06), grade 3 tumors (HR 2.63), and laparoscopic surgery (HR 0.51). For OS, non-endometrioid histology (HR 5.12), stage III disease (HR 2.98), and grade 3 tumors (HR 4.51) were independent adverse factors, whereas surgical approach was not. Conclusions: The use of an intrauterine manipulator in laparoscopic hysterectomy for early-stage endometrial cancer was linked to worse DFS but not OS. These findings suggest caution regarding the routine use of IUMs and highlight the need for prospective randomized trials to clarify their oncological safety. Full article

Automotive manufacturing quality control faces persistent challenges such as limited defect samples, cross-domain variability, and the demand for interpretable decision-making. This work presents an explainable few-shot anomaly detection framework that integrates EfficientNet-based feature extraction, adaptive prototype learning, and component-specific attention mechanisms to address these requirements. The system is designed for rapid adaptation to novel defect types while maintaining interpretability through a multi-modal explainable AI module that combines visual, quantitative, and textual outputs. Evaluation on automotive datasets demonstrates promising performance on evaluated automotive components, achieving 99.4% accuracy for engine wiring inspection and 98.8% for gear inspection, with improvements of 5.2–7.6% over state-of-the-art baselines, including traditional unsupervised methods (PaDiM, PatchCore), advanced approaches (FastFlow, CFA, DRAEM), and few-shot supervised methods (ProtoNet, MatchingNet, RelationNet, FEAT), and with only 0.63% cross-domain degradation between wiring and gear inspection tasks. The architecture operates under real-time industrial constraints, with an average inference time of 18.2 ms, throughput of 60 components per minute, and memory usage below 2 GB on RTX 3080 hardware. Ablation studies confirm the importance of prototype learning (−4.52%), component analyzers (−2.79%), and attention mechanisms (−2.21%), with K = 5 few-shot configuration providing the best trade-off between accuracy and adaptability. Beyond performance, the framework produces interpretable defect localization, root-cause analysis, and severity-based recommendations designed for manufacturing integration with execution systems via standardized industrial protocols. These results demonstrate a practical and scalable approach for intelligent quality control, enabling robust, interpretable, and adaptive inspection within the evaluated automotive components. Full article

This study examines the effects of injection timing and cerium oxide (CeO₂) nanoparticle (NP) size on NOx emissions and brake thermal efficiency (BTE) in a compression ignition engine, contributing to Sustainable Development Goals 7 and 13. Experiments were conducted at four load conditions (25–100%) using NP sizes of 10 nm, 30 nm, and 80 nm. An artificial neural network integrated with multi-objective particle swarm optimization (ANN-PSO) was employed to identify optimal operating parameters. The optimized configurations improved BTE and reduced NOx emissions across all loads; for example, at 75% load, BTE increased from 30.38% (average) to 32.13% (optimum), while simultaneously reducing the NOx emissions from 1322 ppm (average) to 1272 ppm (optimum). Analysis of variance (ANOVA) confirmed load as the most significant factor (p < 0.001), followed by injection timing and NP size. The model predictions closely matched experimental results, validating the optimization approach. The optimization suggests an interpolated optimal NP size of approximately 45 nm, highlighting the potential for further exploration. This integrated experimental and computational approach offers a promising framework for improving combustion efficiency and reducing emissions, thereby advancing cleaner and more sustainable fuel technologies. Full article

The cluster of differentiation 44 (CD44) is a member of the hyaluronic acid (HA) receptor family of cell adhesion molecules. Besides HA, this transmembrane protein also serves as a receptor for other components of the extracellular matrix (ECM), including fibronectin, collagen, and osteopontin (OPN). The CD44-HA axis is involved in a wide range of physiological and cancer-related processes, particularly in cell adhesion and migration, lymphocyte activation, as well as tumour progression and metastasis. The possibility of modulating the CD44-HA interaction with a pharmacological inhibitor has therefore been recognized as an emerging anti-cancer strategy. With its expression in a wide variety, CD44 has also become the most common surface biomarker of cancer stem cells. Due to the rapid progress of research on this crucial receptor, some published and deposited variants were often poorly described or lacked accession numbers in the available protein databases, which created confusion and hindered relevant research. In this work, we attempted to examine the protein sequences of the known CD44 variants and match them between the two UniProt and the National Centre for Biotechnology Information (NCBI) Protein databases. The deposited sequences were aligned to the CD44 canonical sequence and grouped based on the observed differences. Analysis of CD44–ligand experimental structures available in the Protein Data Bank (PDB) was also performed to identify the most promising small-molecule inhibitors of the CD44-HA interaction. Full article