Search Results (3,254)

Objectives: To characterize the learning curve of endoscopic transsphenoidal pituitary adenoma surgery performed by a single neurosurgeon, assess how operative time, resection rates, and clinical outcomes evolved with experience, and identify tumor-related factors influencing surgical performance. Methods: This retrospective study included 123 consecutive endoscopic transsphenoidal pituitary adenoma resections performed between 2018 and 2025. Cases were divided into quartiles according to chronological order. Clinical, radiological, endocrinological, and operative variables were analyzed. Gross total resection (GTR), biochemical remission, postoperative complications, and visual and cranial nerve outcomes were compared between quartiles. A segmented linear regression model was applied to identify changepoints in the operative-time learning curve. Statistical significance was set at p < 0.05. Results: The mean operative time decreased by 31.8%, from 160.8 min in Quartile 1 to 109.7 min in Quartile 4. Segmented regression revealed two changepoints at cases 47 and 85, defining three learning phases: a steep improvement phase, a consolidation phase, and a plateau. GTR was achieved in 51.2% of patients and did not significantly differ across quartiles. For Knosp 0–2 tumors, GTR was 76.1% overall; for Knosp 3–4 tumors, 30%. Tumor diameter, Knosp grade, and sphenoid sinus invasion were strongly associated with lower GTR rates (all p < 0.05). Biochemical remission was achieved in 74.2% of patients with functional adenomas. New or worsened postoperative pituitary insufficiency significantly decreased across quartiles (p < 0.001). Rates of postoperative diabetes insipidus (30.8%) and CSF leak (6.5%) were comparable with published literature and showed no consistent temporal trend. Conclusions: A clear learning curve exists in endoscopic pituitary surgery, with operative proficiency achieved after approximately 50 cases and an experienced plateau after ~90 cases. Surgical experience significantly reduced operative time and postoperative pituitary insufficiency but did not influence GTR rates, likely due to a high and increasing proportion of large tumors with cavernous sinus invasion. Tumor size, Knosp grade, and sphenoid sinus invasion were identified as major determinants of surgical complexity and should be accounted for when evaluating learning curves and surgical outcomes. Full article

The widespread application of robots in complex and dynamic environments demands that Visual SLAM is both robust and accurate. However, dynamic objects, varying illumination, and environmental complexity fundamentally challenge the static world assumptions underlying traditional SLAM methods. This review provides a comprehensive investigation into the mathematical foundations of V-SLAM and systematically analyzes the key optimization techniques developed for dynamic environments, with particular emphasis on advances since 2020. We begin by rigorously deriving the probabilistic formulation of V-SLAM and its basis in nonlinear optimization, unifying it under a Maximum a Posteriori (MAP) estimation framework. We then propose a taxonomy based on how dynamic elements are handled mathematically, which reflects the historical evolution from robust estimation to semantic modeling and then to deep learning. This framework provides detailed analysis of three main categories: (1) robust estimation theory-based methods for outlier rejection, elaborating on the mathematical models of M-estimators and switch variables; (2) semantic information and factor graph-based methods for explicit dynamic object modeling, deriving the joint optimization formulation for multi-object tracking and SLAM; and (3) deep learning-based end-to-end optimization methods, discussing their mathematical foundations and interpretability challenges. This paper delves into the mathematical principles, performance boundaries, and theoretical controversies underlying these approaches, concluding with a summary of future research directions informed by the latest developments in the field. The review aims to provide both a solid mathematical foundation for understanding current dynamic V-SLAM techniques and inspiration for future algorithmic innovations. By adopting a math-first perspective and organizing the field through its core optimization paradigms, this work offers a clarifying framework for both understanding and advancing dynamic V-SLAM. Full article

Individual animal identification is a cornerstone of animal welfare practices and is of crucial importance for food safety and the protection of humans from zoonotic diseases. It is also a key prerequisite for enabling automated processes in modern dairy farming. With newly emerging technologies, visual animal identification based on machine learning offers a more efficient and non-invasive method with high automation potential, accuracy, and practical applicability. However, a common challenge is the limited variability of training datasets, as images are typically captured in controlled environments with uniform backgrounds and fixed poses. This study investigates the impact of foreground segmentation and background removal on the performance of convolutional neural networks (CNNs) for cow identification. A dataset was created in which training images of dairy cows exhibited low variability in pose and background for each individual, whereas the test dataset introduced significant variation in both pose and environment. Both a fine-tuned CNN backbone and a model trained from scratch were evaluated using images with and without background information. The results demonstrate that although training on segmented foregrounds extracts intrinsic biometric features, background cues carry more information for individual recognition. Full article

This study presents a field-oriented AI web system for durian orchard management that recognizes leaf health from on-orchard images under variable illumination. Two complementary pipelines are employed: (1) a rule-based module operating in HSV and CIE Lab color spaces that suppresses sun-induced specular highlights via V/L* thresholds and applies interpretable hue–chromaticity rules with spatial constraints; and (2) a Deep Feature (PCA–SVM) pipeline that extracts features from pretrained ResNet50 and DenseNet201 models, performs dimensionality reduction using Principal Component Analysis, and classifies samples into three agronomic classes: healthy, leaf-spot, and leaf-blight. This hybrid architecture enhances transparency for growers while remaining robust to illumination variations and background clutter typical of on-farm imaging. Preliminary on-farm experiments under real-world field conditions achieved approximately 80% classification accuracy, whereas controlled evaluations using curated test sets showed substantially higher performance for the Deep Features and Ensemble model, with accuracy reaching 0.97–0.99. The web interface supports near-real-time image uploads, annotated visual overlays, and Thai-language outputs. Usability testing with thirty participants indicated very high satisfaction (mean 4.83, SD 0.34). The proposed system serves as both an instructional demonstrator for explainable AI-based image analysis and a practical decision-support tool for digital horticultural monitoring. Full article

Uveitis constitutes a heterogeneous group of intraocular inflammatory pathologies, including both infectious and non-infectious aetiologies, often leading to substantial morbidity and permanent loss of vision in up to 20% of the affected cases. Visual impairment is most prominent in intermediate, posterior, or panuveitis and is commonly associated with cystoid macular oedema, epiretinal membranes, macular holes, and retinal detachment. In the context of uveitis, these complications arise as a result of recurrent flare-ups or chronic inflammation, contributing to cumulative ocular damage. Pars plana vitrectomy (PPV) has an evolving role in the diagnostic and therapeutic approach to uveitis. Diagnostic PPV allows for the analysis of vitreous fluid and tissue using techniques such as PCR, flow cytometry, cytology, and cultures, providing further insights into intraocular immune responses. Therapeutic PPV can be employed for the management of structural complications associated with uveitis, in a wide spectrum of inflammatory clinical entities such as Adamantiades–Behçet disease, juvenile idiopathic arthritis, acute retinal necrosis, or ocular toxoplasmosis. Modern small-gauge and minimally invasive techniques improve visual outcomes, reduce intraocular inflammation, and may decrease reliance on systemic immunosuppression. Emerging technologies, including robot-assisted systems, are expected to enhance surgical precision and safety in the future. Despite these advances, PPV outcomes remain variable due to heterogeneity in indications, surgical techniques, and postoperative management. Prospective studies with standardized protocols, detailed subgroup analyses, and the integration of immunological profiling are needed to define which patients benefit most, optimize therapeutic strategies, and establish predictive biomarkers in uveitis management. Full article

Background: Herpes zoster (HZ) represents a substantial public health concern among aging populations, yet regional variability in clinical patterns and risk determinants remains insufficiently documented. In southeastern Romania, epidemiological data are limited, and the combined influence of demographic, behavioral, and metabolic factors on disease severity has not been systematically evaluated. Methods: We performed a retrospective observational study including 100 consecutive patients diagnosed with HZ between 2019 and 2023 in a dermatology department in southeastern Romania. Demographic characteristics, lifestyle behaviors, anthropometric status, clinical manifestations, and outcomes were extracted from medical records. Associations between categorical variables were assessed using Chi-square tests and Cramer’s V, while interaction patterns were explored through log-linear modeling. Heatmaps were generated in Python (version 3.10) using the Matplotlib library (version 3.7.1) to visualize distribution patterns and subgroup relationships. Results: The cohort showed a marked age dependence, with 77% of cases occurring in individuals ≥ 60 years, consistent with immunosenescence-driven reactivation. Women represented 59% of cases, and 84.7% of female patients were postmenopausal. Urban residents predominated (91%). Vesicular eruption (84%) and acute pain (79%) were the most frequent symptoms. Localized HZ was observed in 81% of cases, while ophthalmic involvement (11%) and disseminated forms (8%) were less common. Lifestyle factors significantly influenced clinical severity: smokers, alcohol consumers, and sedentary individuals exhibited higher proportions of postherpetic neuralgia (PHN) and ocular complications (p < 0.001). Overweight and obese patients demonstrated a higher burden of PHN, suggesting a role for metabolic inflammation, although BMI was not associated with incidence. No significant association between age category and complication type was detected, likely due to small subgroup sizes despite a clear descriptive trend toward increased severity with advanced age. Conclusions: These findings support a multifactorial model of HZ severity in southeastern Romania, shaped by age, lifestyle behaviors, hormonal status, and metabolic risk. While incidence patterns align with international data, the strong impact of modifiable factors on complication rates highlights the need for targeted prevention and individualized risk assessment. Results offer a regional perspective that may inform future multicenter investigations. Full article

Smart packaging is an alternative that may not only replace plastic containers, but also enable food quality monitoring. In this study, an innovative packaging system was developed using a starch-chitosan polymer matrix, infused with rosemary essential oil (REO) as an antimicrobial agent, and betalain extract as a food quality indicator. Betalain extract, derived from beet waste, can change color with pH, making it a useful natural indicator for monitoring food freshness. This packaging system is beneficial for foods that produce metabolites related to degradation, which alter pH and allow for the visual detection of changes in product quality. The objective of this work was to develop a smart packaging system with betalains and rosemary essential oil (REO) to extend food shelf life and monitor quality during storage. REO demonstrated antimicrobial activity, but its effect did not differ significantly among the microorganisms tested. On the other hand, the betalain extract (35.75% BE v/v) completely inhibited the growth of Listeria innocua and Salmonella spp. at concentrations of 50% (v/v; 0.82 ± 0.04 mg betalain/g), showing its potential as an antimicrobial agent. The interactions between chitosan and betalains were primarily associated with electrostatic interactions between the positively charged amino groups of chitosan and the negatively charged carboxyl groups of betalains. In contrast to starch, these interactions could result from interactions between the C=O groups of betalain carboxyls and water, which, in turn, interact with the hydroxyl groups of starch through hydrogen bonding. Despite the results obtained in this study, certain limitations need to be addressed in future research, such as the variability in antimicrobial activity among different bacterial strains, which could reveal differences in the efficacy of betalains and essential oils against other pathogens. Full article

With the rapid advancements in technology, there has been a noticeable increase in phishing attacks that exploit users by impersonating trusted entities. The primary attack vectors include fraudulent websites and carefully crafted emails. Early detection of such threats enables the more effective blocking of malicious sites and timely user warnings. One of the key elements in phishing detection is identifying the entity being impersonated. In this article, we conduct a comparative analysis of methods for detecting phishing websites that rely on website screenshots and recognizing their impersonation targets. The two main research objectives include binary phishing detection to identify malicious intent and multiclass classification of impersonated targets to enable specific incident response and brand protection. Three approaches are compared: two state-of-the-art methods, Phishpedia and VisualPhishNet, and a third, proposed in this work, which uses perceptual hash similarity as a baseline. To ensure consistent evaluation conditions, a dedicated framework was developed for the study and shared with the community via GitHub. The obtained results indicate that Phishpedia and the Baseline method were the most effective in terms of detection performance, outperforming VisualPhishNet. Specifically, the proposed Baseline method achieved an F1 score of 0.95 on the Phishpedia dataset for binary classification, while Phishpedia maintained a high Identification Rate (>0.9) across all tested datasets. In contrast, VisualPhishNet struggled with dataset variability, achieving an F1 score of only 0.17 on the same benchmark. Moreover, as our proposed Baseline method demonstrated superior stability and binary classification performance, it should be considered as a robust candidate for preliminary filtering in hybrid systems. Full article

Objective: This study investigates the association between consumers’ perceived readability of Multiple Traffic Light (MTL) label print size—a theoretical structural gatekeeper for visual salience—and self-reported food consumption frequency in the United Kingdom. We aimed to disentangle the effect of label readability from label content. Using non-equivalent dependent variables (NEDVs), we tested whether the association is specific to unhealthy convenience foods and absent for healthy or unlabeled foods, while also examining heterogeneity across consumer subgroups. Methods: Data from 8948 adults across four waves (2012–2018) of the UK Food and You Survey were analyzed. Cumulative link ordinal logistic regressions were employed to model the association between self-reported print size readability and the consumption frequency of four product types: pre-packaged sandwiches and pre-cooked meat (unhealthy, labeled targets), dairy (nutritionally advisable, labeled control), and fresh meat (unlabeled control). Models were adjusted for sociodemographic covariates, health behaviors, and survey wave fixed effects. Results: The findings reveal a content-specific and significant dynamic relationship exclusively for pre-packaged sandwiches. In 2012, a one-unit increase in readability was associated with a 9% decrease in the odds of frequent consumption ($OR=0.91$), consistent with a warning effect. However, by 2018, this relationship reversed to a 4% increase ($OR=1.04$), indicating that higher readability became associated with more frequent consumption. In contrast, a persistent null association was observed for pre-cooked meat, dairy, and fresh meat. Subgroup analyses for sandwiches indicated that the association with readability was strongest among less-engaged consumers. Conclusions: Empirical evidence challenges the utility of a standardized approach to food labelling. The results suggest that the effectiveness of label salience is contingent not just on the consumer but on the product’s context and the content of its message, highlighting the need for adaptive rather than uniform policy standards. Full article

Quick Quantitative (QQ) immunoassays have been increasingly applied for the measurement of enrofloxacin (ENR) and ciprofloxacin (CIP) residues in aquaculture due to their speed and convenience. However, their quantitative reliability remains limited because measurement uncertainty (MU) is rarely considered during field testing. To enhance the metrological reliability of QQ-based residue analysis, we developed AquaUncertainty Pal, a mobile application that embeds real-time MU computation into the QQ workflow. The software automatically evaluates uncertainty sources during sampling and pipetting, visualizes the uncertainty budget, and guides users through optimized operations. The framework was validated against ISO/IEC 17025–accredited LC–MS/MS and assessed through a user study involving 20 frontline technicians. With the integrated software, pipetting precision (RSD) at 100 μL improved from 4.1% to 1.79%, the inter-operator variability (CV) decreased by 52%, and conformity assessment accuracy for samples near the maximum residue limit (MRL) increased from 25% to 70%. This suggests that real-time MU visualization effectively guided technicians toward consistent pipetting and interpretation behavior. These results demonstrate that integrating MU into the QQ workflow is both feasible and effective, substantially improving reliability and providing a replicable digital framework for uncertainty-informed residue monitoring in aquaculture. Full article

This study develops an immersive marine Information and Communication Technology (ICT) convergence framework designed to enhance coastal climate resilience by improving accessibility, visualization, and communication of scientific research on Dokdo (Dok Island) in the East Sea. High-resolution spatial datasets, multi-source marine observations, underwater imagery, and validated research outputs were integrated into an interactive virtual-reality (VR) and web-based three-dimensional (3D) platform that translates complex geophysical and ecological information into intuitive experiential formats. A geospatially accurate 3D virtual model of Dokdo was constructed from maritime and underwater spatial data and coupled with immersive VR scenarios depicting sea-level variability, coastal morphology, wave exposure, and ecological characteristics. To evaluate practical usability and pro environmental public engagement, a three-phase field survey (n = 174) and a System Usability Scale (SUS) assessment (n = 42) were conducted. The results indicate high satisfaction (88.5%), strong willingness to re-engage (97.1%), and excellent usability (mean SUS score = 80.18), demonstrating the effectiveness of immersive content for environmental education and science communication crucial for achieving Sustainable Development Goal 14 targets. The proposed platform supports stakeholder engagement, affective learning, early climate risk perception, conservation planning, and multidisciplinary science–policy dialogue. In addition, it establishes a foundation for a digital twin system capable of integrating real-time ecological sensor data for environmental monitoring and scenario-based simulation. Overall, this integrated ICT-driven framework provides a transferable model for visualizing marine research outputs, enhancing public understanding of coastal change, and supporting sustainable and adaptive decision-making in small island and coastal regions. Full article

This study provides a multi-faceted and detailed perspective on the relationships between logistics performance, environmental degradation, and economic growth in 38 OECD countries, using each as an individual target variable. In the Analysis section, the relationship between logistics and environment is examined within a broader context, taking economic indicators into account. This examination utilizes the machine learning algorithms Random Forest (RF), Extreme Gradient Boosting (XGBoost), and Light Gradient Boosting Machine (LightGBM). For each algorithm, the dataset is split into training and testing sets using three different ratios: 90:10, 80:20, and 70:30. A comprehensive performance evaluation is conducted on each of these splits by applying 5-fold and 10-fold cross-validation (CV). Considering economic indicators, the analysis section examines how the logistics-environment relationship is shaped in a broader context using the machine learning algorithms RF, XGBoost, and LightGBM. MSE, MAE, RMSE, MAPE, and R² metrics are utilized to evaluate model performance, while MDA and SHAP are employed to assess feature importance. Furthermore, a bee swarm plot is leveraged for visualizing the results. The XGBoost algorithm can successfully predict carbon dioxide (CO₂) emissions from transport and economic growth with high accuracy. However, the logistics performance model achieves high performance only with the LightGBM algorithm using a 90% train, 10% test split, and 5-fold CV setup. Based on the variable importance levels of the best-performing algorithm for each of the three target variables separately, the prediction of logistics performance is largely dependent on the economic growth predictor, and secondly, on the trade openness predictor. In predicting CO₂ emissions from transport, economic growth is identified as the most effective predictor, while logistics performance and trade openness contribute the least to the prediction. The findings also reveal that transport-related emissions and environmental indicators are prominent in the prediction of economic growth, whereas logistics performance and trade openness play a supportive, yet secondary role. Full article

Background: Autism spectrum disorder (ASD) is associated with distinct visual attention patterns that provide insight into underlying social-cognitive mechanisms. Methods: This systematic review (PROSPERO: CRD42023429316), conducted per PRISMA guidelines, synthesizes evidence from 14 peer-reviewed studies using eye-tracking to compare oculomotor strategies in autistic children and typically developing (TD) controls. A comprehensive literature search was conducted in PubMed, Web of Science, and Science Direct up to March 2025. Study inclusion criteria focused on ASD versus TD group comparisons in individuals under 18 years, with key metrics, fixation duration and count, spatial distribution, saccadic parameters systematically extracted. Risk of bias was assessed using the QUADAS-2 tool, revealing high heterogeneity in both index tests and patient selection. Results: The results indicate that autistic children exhibit reduced fixation on socially salient stimuli, atypical saccadic behavior, and more variable spatial exploration compared to controls. Conclusions: These oculomotor differences suggest altered mechanisms of social attention and information processing in ASD. Findings suggest that eye-tracking can contribute valuable information about heterogeneous gaze profiles in ASD, providing preliminary insight that may inform future studies to develop more sensitive diagnostic tools. This review highlights visual attention patterns as promising indicators of neurocognitive functioning in ASD. Full article

Plasma cell-free RNA (cfRNA) metagenomics is increasingly explored for blood-based pathogen detection, but the structure of the shared background “blood microbiome”, the reproducibility of reported signals, and the practical limits of this approach remain unclear. We performed a critical re-analysis and benchmarking (“stress test”) of host-filtered blood RNA sequencing data from two cohorts: a bacteriologically confirmed tuberculosis (TB) cohort (n = 51) previously used only to derive host cfRNA signatures, and a coronary artery disease (CAD) cohort (n = 16) previously reported to show a CAD-shifted “blood microbiome” enriched for periodontal taxa. Both datasets were processed with a unified pipeline combining stringent human read removal and taxonomic profiling using the latest versions of specialized tools Kraken2 and MetaPhlAn4. Across both cohorts, only a minority of non-host reads were classifiable; under strict host filtering, classified non-host reads comprised 7.3% (5.0–12.0%) in CAD and 21.8% (5.4–31.5%) in TB, still representing only a small fraction of total cfRNA. Classified non-host communities were dominated by recurrent, low-abundance taxa from skin, oral, and environmental lineages, forming a largely shared, low-complexity background in both TB and CAD. Background-derived bacterial signatures showed only modest separation between disease and control groups, with wide intra-group variability. Mycobacterium tuberculosis-assigned reads were detectable in many TB-positive samples but accounted for ≤0.001% of total cfRNA and occurred at similar orders of magnitude in a subset of TB-negative samples, precluding robust discrimination. Phylogeny-aware visualization confirmed that visually “enriched” taxa in TB-positive plasma arose mainly from background-associated clades rather than a distinct pathogen-specific cluster. Collectively, these findings provide a quantitative benchmark of the background-dominated regime and practical limits of plasma cfRNA metagenomics for pathogen detection, highlighting that practical performance is constrained more by a shared, low-complexity background and sparse pathogen-derived fragments than by large disease-specific shifts, underscoring the need for transparent host filtering, explicit background modeling, and integration with targeted or orthogonal assays. Full article

Detecting vehicles in remote sensing images of construction sites captured by Unmanned Aerial Vehicles (UAVs) faces severe challenges, including extremely small target scales, high inter-class visual similarity, cluttered backgrounds, and highly variable imaging conditions. To address these issues, we propose BSGNet (Biomimetic Sharpening and Global Receptive Field Network)—a novel detection architecture that synergistically fuses biologically inspired visual mechanisms with global receptive field modeling. Inspired by the Sustained Contrast Detection (SCD) mechanism in frog retinal ganglion cells, we design a Perceptual Sharpening Module (PSM). This module combines dual-path contrast enhancement with spatial attention mechanisms to significantly improve sensitivity to the high-frequency edge structures of small targets while effectively suppressing interfering backgrounds. To overcome the inherent limitation of such biomimetic mechanisms—specifically their restricted local receptive fields—we further introduce a Global Heterogeneous Receptive Field Learning Module (GRM). This module employs parallel multi-branch dilated convolutions and local detail enhancement paths to achieve joint modeling of long-range semantic context and fine-grained local features. Extensive experiments on our newly constructed UAV Construction Vehicle (UCV) dataset demonstrate that BSGNet achieves state-of-the-art performance: obtaining 64.9% APs on small targets and 81.2% on the overall mAP@0.5 metric, with an inference latency of only 31.4 milliseconds, outperforming existing mainstream detection frameworks in multiple metrics. Furthermore, the model demonstrates robust generalization performance on public datasets. Full article