Search Results (794)

This paper presents an extended investigation into deep learning-based cryptanalysis of block ciphers by introducing and evaluating a multi-server attack environment. Building upon our prior work in centralized settings, we explore the practicality and scalability of deploying such attacks across multiple distributed edge servers. We assess the vulnerability of five representative block ciphers—DES, SDES, AES-128, SAES, and SPECK32/64—under two neural attack models: Encryption Emulation (EE) and Plaintext Recovery (PR), using both fully connected neural networks and Recurrent Neural Networks (RNNs) based on bidirectional Long Short-Term Memory (BiLSTM). Our experimental results show that the proposed federated learning-based cryptanalysis framework achieves performance nearly identical to that of centralized attacks, particularly for ciphers with low round complexity. Even as the number of edge servers increases to 32, the attack models maintain high accuracy in reduced-round settings. We validate our security assessments through formal statistical significance testing using two-tailed binomial tests with 99% confidence intervals. Additionally, our scalability analysis demonstrates that aggregation times remain negligible (<0.01% of total training time), confirming the computational efficiency of the federated framework. Overall, this work provides both a scalable cryptanalysis framework and valuable insights into the design of cryptographic algorithms that are resilient to distributed, deep learning-based threats. Full article

Background/Objectives: Veterinarians are among the most stressed of all professional groups. Their work is characterised by long working hours, high emotional demands and an increased risk of anxiety, depression, suicide and burnout. The aim of this cross-sectional study that examines retrospective records of experienced violence was to analyse the frequency of violent acts and their connection to certain factors (age, gender, place of work, and specialist area according to animal species). Methods: This nationwide, cross-sectional, online survey of veterinarians in Germany was conducted between July 2021 and February 2023. A total of 1053 veterinarians were included in the analysis, which was conducted according to the respondents’ age, (<35 years, 35–45 years and >45 years), gender, workplace and veterinary specialisation. Sociodemographic and work-related data were collected, as were responses to questions regarding experiences of violence, which were differentiated between verbal abuse and physical violence. The data were analysed using descriptive statistics and non-parametric group comparisons (Kruskal–Wallis test with Bonferroni correction, Mann–Whitney U test and Pearson’s chi-squared test). Results: Overall, 52.7% of veterinarians reported experiencing verbal abuse or physical violence at the hands of animal owners. Verbal abuse occurred, on average, more than three times per month, whereas physical violence was rare. Physical violence occurred significantly more frequently among middle-aged veterinarians (p < 0.001). The highest prevalence of verbal abuse or violence (72.5%, p < 0.001) was reported by veterinarians working in public authorities, while the lowest was reported by those working in laboratories. Conclusions: Workplace violence against veterinarians is a frequent occupational burden in Germany and highlights the urgent need for targeted prevention, de-escalation training and organisational support across veterinary settings. Full article

The present exploratory study investigates how pre-primary and primary student teachers (STs) at the Universidad Autónoma de Madrid (UAM) perceive the impact of bilingual education programs (BEPs) on children’s learning experience. Specifically, it examines student teachers’ views on the benefits and challenges of implementing Content and Language Integrated Learning (CLIL) in pre-primary and primary education and explores whether there are differences between the opinions of the two groups. The analysis is based on data from six items of a structured questionnaire, validated in previous studies and completed by 170 prospective pre-primary and primary teachers at the UAM. The results suggest a shared perception among STs that BEPs enrich the learning experience of students in both pre-primary and primary education. Most STs recognize that CLIL enhances language proficiency and supports cognitive development, although they also point to insufficient teacher training and the low motivation of children with learning difficulties as major challenges. While no major differences emerged between the views of pre-primary and primary STs, subtle variations point to the existence of two distinct trainee profiles that determine their views on BEPs and that would require further mid-term investigation. The findings highlight areas for targeted support in teacher training programs. Full article

Background/Objectives: Endoscopic submucosal dissection (ESD) is a state-of-the-art en bloc resection for early gastro-intestinal cancers and precursors developed and validated in Japan. Western expertise with this complex technique remains limited. Tutored training might be optimal for patients and ESD learning. We established ESD tutoring courses led by experienced Japanese experts to provide (i) optimal long-term curative outcomes and low complication rates for patients and (ii) hands-on training on difficult lesions for European endoscopists under direct expert supervision. Methods: Prospective data from 2011 to 2015 (follow-up to 12/2024) were analyzed. A total of 118 neoplasms (50% HGIEN and cancer) in 101 patients (median age 68 [37–91] years; 38% with significant comorbidities) were treated with expert or tutored ESD. Japanese experts performed 28 ESDs, while 22 trained beginners conducted 90 supervised procedures on difficult lesions during 5 live and 20 tutoring events (1–4 days each). Results: Analysis of the complete data showed curative and en bloc resection rates of 88% and 95%, respectively, with no recurrence after R0 resections during a median follow-up of 9.8 [1.5–14.9] years. Long-term survival remained recurrence-free after endoscopic resection of 3 recurrent adenomas (at R1/Rx) and curative surgery/2nd ESD for 5 non-curative ESDs. Adverse events occurred in 9.3% without emergency surgery or 30-day mortality. Comparing expert-only vs. tutored ESD procedures, beginners correctly applied curative ESD indications in 94% of 118 neoplasms. Experts resected larger lesions (22 cm²) at a rate of 9.3 cm²/h in 121 min. Tutored beginners achieved a 75% [25–100] self-completion rate on 33% smaller lesions in 112 min. Conclusions: ESD tutoring courses led by Japanese experts ensure excellent patient outcomes and standardized procedural training. This model may foster professional ESD performance across European referral centers. Full article

Bio-electric fields—manifested as Electroencephalogram (EEG), Electrocardiogram (ECG), Electromyogram (EMG), and Electrooculogram (EOG)—are fundamental to modern medical diagnostics but often suffer from severe data imbalance, scarcity, and environmental noise. Generative Adversarial Networks (GANs) offer a powerful, nonlinear solution to these modeling hurdles. This review presents a comprehensive survey of GAN methodologies specifically tailored for bio-electric signal processing. We first establish a theoretical foundation by detailing GAN principles, training mechanisms, and critical structural variants, including advancements in loss functions and conditional architectures. Subsequently, the paper extensively analyzes applications ranging from high-fidelity signal synthesis and noise reduction to multi-class classification. Special attention is given to clinical anomaly detection, specifically covering epilepsy, arrhythmia, depression, and sleep apnea. Furthermore, we explore emerging applications such as modal transformation, Brain–Computer Interfaces (BCI), de-identification for privacy, and signal reconstruction. Finally, we critically evaluate the computational trade-offs and stability issues inherent in current models. The study concludes by delineating prospective research avenues, emphasizing the necessity of interdisciplinary synergy to advance personalized medicine and intelligent diagnostic systems. Full article

To address the application requirements of Visual Simultaneous Localization and Mapping (VSLAM) in intracavitary environments and the scarcity of gold-standard datasets for deep learning methods, this study proposes a hybrid attention mechanism combined with U-Net for vascular branch point extraction in endoluminal images (SuperVessel). The network is initialized via transfer learning with pre-trained SuperRetina model parameters and integrated with a vascular feature detection and matching method based on dual branch fusion and structure enhancement, generating a pseudo-gold-standard vascular branch point dataset. The framework employs a dual-decoder architecture, incorporates a dynamic up-sampling module (CBAM-Dysample) to refine local vessel features through hybrid attention mechanisms, designs a Dice-Det loss function weighted by branching features to prioritize vessel junctions, and introduces a dynamically weighted Triplet-Des loss function optimized for descriptor discrimination. Experiments on the Vivo test set demonstrate that the proposed method achieves an average Area Under Curve (AUC) of 0.760, with mean feature points, accuracy, and repeatability scores of 42,795, 0.5294, and 0.46, respectively. Compared to SuperRetina, the method maintains matching stability while exhibiting superior repeatability, feature point density, and robustness in low-texture/deformation scenarios. Ablation studies confirm the CBAM-Dysample module’s efficacy in enhancing feature expression and convergence speed, offering a robust solution for intracavitary SLAM systems. Full article

Soil organic carbon (SOC) is an essential indicator of soil fertility, health, and carbon sequestration capacity. Its proper management improves soil structure, productivity, and resilience to climate change, making rapid and reliable SOC assessment essential for sustainable agriculture. Visible and near-infrared (Vis–NIR) spectroscopy offers a non-destructive and cost-effective alternative to conventional laboratory analyses, allowing for the simultaneous estimation of multiple soil properties from a single spectrum. This study aimed to predict SOC content using machine learning techniques applied to Vis–NIR spectra of 860 soil samples collected in the Sierra Nevada de Santa Marta, Colombia. The spectra (400–2500 nm) were acquired using a NIR spectrophotometer, and the soil organic carbon (SOC) content was quantified using a wet oxidation method that employs dichromate in an acidic medium. A hybrid modeling framework combining Random Forest (RF) with support vector regression (SVR) and XGBoost was implemented. Spectral pretreatments (Savitzky–Golay first derivative, MSC, and SNV) were compared, and spectral bands were selected every 10 nm. The 30 most relevant wavelengths were identified using RF importance analysis. Data were divided into training (80%) and test (20%) subsets using stratified random sampling, and five-fold cross-validation was applied for parameter optimization and overfitting control. The RF–XGBoost (R² = 0.86) and RF–SVR (R² = 0.85) models outperformed the individual RF and SVR models (R² < 0.7). The proposed hybrid approach, optimized through features, and advanced spectral preprocessing demonstrate a robust and scalable framework for rapid prediction of SOC and sustainable soil monitoring. Full article

Reservoir petrophysical characterization is traditionally performed through the interpretation of well logs validated with routine core analysis (RCAL), often excluding the integration of other tools such as computed tomography (CT), which provides interpretation of higher resolution. In this study, artificial neural network (ANN) models were applied to estimate porosity and permeability by integrating conventional logs with CT-derived data (RHOB and PEF), thereby validating the petrophysical model of Ciénaga de Oro Formation. Neural networks were trained in MATLAB^® using a feed-forward regression network based on a multilayer perceptron (MLP) architecture, with RCAL measurements serving as a reference. Model performance was assessed by comparing predictions with laboratory data from two wells, yielding high accuracy (R² = 0.98 for permeability and R² = 0.90 for porosity) with mean absolute errors below 5%. Additional validation was performed using well logs and CT data from complete 3 ft sections, with the trained models successfully reproducing core heterogeneities at millimetric resolution. These results confirm the potential of integrating well logs and CT data with ANN to enhance petrophysical characterization and extend property estimation to wells lacking core or laboratory measurements. Furthermore, an interactive MATLAB^® tool was developed, enabling users to load well logs and CT files as flat inputs, generate high-resolution predictions, validate results, and export the estimated values. Full article

Optimizing coagulant dosages in Drinking Water Treatment Plants (DWTPs) is critical for reducing operational costs, minimizing chemical waste, mitigating environmental impacts, and ensuring consistent water quality, particularly in resource-constrained settings where conventional jar tests are labor-intensive and poorly suited to real-time demands. This study develops and validates a Random Forest (RF) machine learning model to predict optimal dosages of aluminum sulfate, polyaluminum chloride, and a polymer flocculant at the Miguel de la Cuba Ibarra DWTP in Peru, addressing the need for an efficient, real-time decision support system. Using a historical dataset of 2556 jar tests, a univariate RF model was developed to predict settled water turbidity, tailored to the plant’s typical operational range. The model demonstrated robust predictive performance, achieving a coefficient of determination (R²) of 0.92 during training and 0.76 during validation with unseen data, alongside a Root Mean Square Error (RMSE) of 0.11 NTU and a Mean Absolute Percentage Error (MAPE) of 0.11 in the training phase. Integrated into a digital platform, the model generates real-time NTU ppm dosing curves, providing a practical and responsive tool to enhance operational efficiency for DWTP operators. This work offers a scalable, data-driven solution to improve water treatment processes in resource-limited contexts. Full article

After the Spanish Civil War, the shortage of building materials in the country and the restrictions imposed by the Dirección General de Arquitectura limited the use of steel in construction, encouraging solutions that reduced the consumption of this material. In this context, the thin-tile vault gained new relevance due to its low cost, speed of execution and good structural and fire performance. Among the architects who revisited this system, Ignasi Bosch Reitg (1910–1985) developed an innovative procedure for the construction of continuous ceilings, based on double-curved vaults with a single layer of brick. His cousin, Josep Maria Bosch Aymerich (1917–2015), an industrial engineer and architect trained in the United States, brought a business vision to the table when he discovered the potential of this system. This paper proposes an in-depth study of the patents requested on this system by the two architects, questioning the reasons for their success or failure in different countries, both in terms of dissemination and exploitation, in regard to the historical context in which it was developed. The analysis, based on original documents from the Bosch Aymerich Archive, uncovers the tensions that the reinterpretation and global projection of a traditional technique can generate. Full article

Background/Objectives: Dental training is known for its demanding academic pace, early clinical exposure, and constant performance pressure. These stressors may contribute to behavioral and physical manifestations, including bruxism and temporomandibular disorder (TMD). This study aimed to better understand the multidimensional burden experienced in this educational setting by investigating the relationship between education-related stress, bruxism patterns, and temporomandibular symptoms (TMD-related symptoms) in dental students. Methods: A cross-sectional survey was conducted at the Ankara University Faculty of Dentistry in 2025 and completed by 287 undergraduate dental students. The questionnaire collected sociodemographic information, self-reported bruxism status, TMD-related symptoms via the Fonseca Anamnestic Index (FAI), and education-related stressors using the Dental Environment Stress (DES) scale. Descriptive statistics, group comparisons, and Spearman correlation analyses were conducted. Results: Bruxism was reported by 76% of students and was significantly more common among females (p < 0.05). Students with bruxism demonstrated higher DES (3.34 ± 0.84) and FAI (41.81 ± 20.32) scores compared with those without bruxism (p < 0.001). DES and FAI scores showed a significant positive correlation (r = 0.229, p < 0.001). Stressors related to workload, examinations, limited rest time, clinical uncertainty, patient responsibility, and financial concerns were strongly associated with bruxism, while inconsistent academic feedback emerged as a key distinguishing factor. Conclusions: Education-related stress is closely linked to bruxism and TMD-related symptoms among dental students. Beyond overall stress intensity, the nature of experienced stressors plays a critical role. These findings highlight the importance of supportive learning structures, targeted stress-management strategies, and curriculum-level improvements to promote student wellbeing and resilience. Full article

Financial statement fraud remains a substantial risk in environments marked by weak regulatory oversight and information asymmetry. This study develops a decision-centric framework that integrates machine learning, explainable artificial intelligence, and decision curve analysis to improve fraud detection under severe class imbalance. Using 969 firm-year observations from 132 Mongolian firms (2013–2024), we evaluate 21 financial ratios with models including Random Forest, XGBoost, LightGBM, MLP, TabNet, and a Stacking Ensemble trained with SMOTE and class-weighted learning. Performance was assessed using PR-AUC, F1-score, Recall, and DeLong-based significance testing. The Stacking Ensemble achieved the strongest results (PR-AUC = 0.93; F1 = 0.83), outperforming both classical and modern baseline models. Interpretability analyses (SHAP, LIME, and counterfactual explanations) consistently identified leverage, profitability, and liquidity indicators as dominant drivers of fraud risk, supported by a SHAP Stability Index of 0.87. Decision curve analysis showed that calibrated thresholds improved decision efficiency by 7–9% and reduced over-audit costs by 3–4%, while an audit cost simulation estimated annual savings of 80–100 million MNT. Overall, the proposed ML–XAI–DCA framework offers a transparent, interpretable, and cost-efficient approach for enhancing fraud detection in emerging-market contexts with limited textual disclosures. Full article

Background/Objectives: Inborn errors of metabolism (IEM) are chronic, life-threatening genetic disorders with a significant cumulative prevalence worldwide. Advances in early diagnosis and treatment have significantly increased life expectancy, underscoring the need for specialised adult care units and the establishment of structured transition programmes from paediatric to adult services. We hereby present a functional transition model for IEM patients and share our implementation experience. Methods: Initiated in 2012, the partnership between the paediatric Hospital Sant Joan de Déu (HSJD) and the adult-care centre at Hospital Clinic of Barcelona (HCB) culminated in 2019 with the transference of the first IEM patients under the structured A10! Programme. This model is structured around the transition units of paediatric and adult centres to guarantee communication and functional management. Regular monthly meetings at each centre and joint quarterly sessions allowed for protocol harmonisation and personalised care planning. Coordinated engagement of the multidisciplinary health care teams with patients and families smoothed the transfer process. Results: Between 2019 and 2024, 94 IEM patients were successfully transferred. Diagnoses included intermediary metabolism defects (71.23%), lipid metabolism and transport disorders (4.25%), heterocyclic compound metabolism (2.12%), complex molecules and organelle dysfunction (6.37%), cofactor and mineral metabolism (2.12%), signalling defects (5.31%), and unclassified cases (8.51% of rare disorders, maybe non-IEM). Transition formats included 21 in-person joint visits in HSJD, 37 remote transitions during the COVID-19 pandemic, and 36 streamlined transfers via standardised protocols. Sessions, trainings, and meetings allowed the exchange of patients’ needs and protocols. Conclusions: The successful transference of IEM patients requires structured programmes with interdisciplinary paediatric and adult teams, joining efforts with the patient, families, and caregivers. Communication between paediatric and adult transition units is essential to promote continuity of care and patient empowerment. While constantly updated, this model has proven effective, gaining positive evaluations from healthcare professionals and patients alike, representing a scalable framework for lifelong management of IEM in adult care settings. Full article

The Réseau de transport de la Capitale (RTC), Quebec City’s public transportation provider, has launched a training program to enhance skills and self-efficacy for using the bus, including training for people with disabilities: “Service d’accompagnement en mobilité intégrée (SAMI)”. This pre-post study with a convergent mixed approach aimed to evaluate the influence of the SAMI program (P-SAMI) on transportation self-efficacy, mobility and satisfaction with the bus use among people with disabilities. The study also explored people with disabilities’ experiences and perceptions with the P-SAMI and bus use. The P-SAMI was delivered, and questionnaires and semi-structured interviews were completed before and after P-SAMI. Paired t-tests, Wilcoxon tests, and deductive thematic analyses were performed. Thirty-three participants (53.7 ± 14.9 years-of-age) showed statistically significant gains in transportation self-efficacy (p < 0.01) and satisfaction with bus use (p < 0.01), with no statistically significant differences in mobility (p > 0.05). Qualitative findings confirmed enhanced transportation self-efficacy and satisfaction with bus use, with participants reporting using buses to carry out some daily activities. The P-SAMI shows promise for improving transportation self-efficacy and satisfaction with using the bus, with the potential to enhance participation in daily activities. Controlled trials should be conducted in the future to test the effectiveness of transportation training for people with disabilities. Full article

Background: Population aging has led to a rise in dementia prevalence, increasing the demand for innovative care models. Telemedicine offers an opportunity to improve access, continuity, and caregiver support for older adults with cognitive impairment. Methods: This qualitative descriptive study was conducted at the Geriatrics and Gerontology Service of Cassiano Antônio de Moraes University Hospital (HUCAM-UFES), Brazil. Semi-structured interviews were carried out with 11 caregivers of older adults living with dementia who participated in telemedicine consultations. Data was analyzed thematically using a reflexive thematic analysis approach. Results: Caregivers considered telemedicine useful, accessible, and safe, facilitating the continuity of care and strengthening the caregiver–professional relationship. The main limitations were the absence of physical examination and occasional technical difficulties. Most caregivers favored a hybrid care model, combining remote and in-person visits. Conclusions: Telemedicine proved to be a feasible and well-accepted strategy for the care of older adults with dementia, improving caregiver support and communication with healthcare teams. Public policies should foster digital inclusion and training for both caregivers and professionals, consolidating hybrid, person-centered models of care. Full article