Search Results (14,235)

The growing global population has intensified the search for sustainable and nutritious food sources. This study evaluated the effects of incorporating giant mealworm (Zophobas morio) flour at levels of 0%, 7.5%, 15%, and 17.5% on the nutritional, physicochemical, and technological properties of fresh tagliatelle pasta. The formulations were characterized in terms of proximate composition, caloric value, amino acid profile, cooking quality, texture, color, microstructure, and structural organization. Protein content increased from 8.16% in the control to 16.29% at the highest enrichment level, while crude fiber rose from 0.25% to 1.30%. The amino acid profile revealed seven essential amino acids, with branched-chain amino acids representing approximately 50% of the essential fraction. Enriched pasta showed a significant reduction in cooking solid loss, decreasing from 0.61 g to 0.21 g, while volume increase remained unchanged. Texture analysis indicated reduced hardness (12.59 to 10.05 gf) and chewiness (5.29 to 2.93 gf), resulting in a softer and less adhesive product, without significant changes in elasticity. Although lightness decreased, the visual appearance was comparable to whole-grain pasta products. Overall, Zophobas morio flour improved the nutritional profile and technological performance of fresh pasta, supporting its use as a sustainable and functional protein ingredient. Full article

Hand localization in cluttered industrial environments remains challenging due to variations in appearance and the gap between synthetic and real-world data. Domain randomization addresses this “reality gap” by intentionally introducing randomized and unrealistic visual features in simulated scenes, encouraging neural networks to focus on essential domain-invariant cues. In this study, we applied domain randomization to generate a synthetic Red-Green-Blue–Depth (RGB-D) dataset for training multimodal instance segmentation models, with the aim of achieving color-agnostic hand localization in complex industrial settings. We introduce a new synthetic dataset tailored to various hand detection tasks and provide ready-to-use pretrained instance segmentation models. To enhance robustness in unstructured environments, the proposed approach employs multimodal inputs that combine color and depth information. To evaluate the contribution of each modality, we analyzed the individual and combined effects of color and depth on model performance. All evaluated models were trained exclusively on the proposed synthetic dataset. Despite the absence of real-world training data, the results demonstrate that our models outperform corresponding models trained on existing state-of-the-art datasets, achieving higher Average Precision and Probability-Based Detection Quality. Full article

This study investigates how two widely used language-modelled generative AI tools, ChatGPT-5.1 (with DALL·E 3) and Gemini 3 (with Imagen), perform as early-stage co-design partners for biophilic interior design. Focusing on real-world use rather than theoretical capability, the research asks to what extent these systems can generate conceptually robust, visually coherent and practically feasible proposals when designers explicitly request biophilic strategies. A multiple-case design was employed across three scenarios: (1) an empty “tabula rasa” room, (2) a damaged rustic room requiring contextual renovation, and (3) a hospital staff break room to be transformed into a “cognitive restoration sanctuary.” For each case, both tools were prompted to produce a step-by-step biophilic design plan and a corresponding photorealistic image. Textual outputs were coded against the 14 Patterns of Biophilic Design and related restorative concepts, while images were evaluated by an expert panel of 15 architects with formal training in biophilic design using a structured Likert-scale instrument. Exterior and building-scale applications were not assessed. Results show that both systems can articulate broadly plausible biophilic strategies but differ in emphasis: ChatGPT tends to produce more spatially coherent, pattern-rich and functionally grounded plans, whereas Gemini excels more in visual realism and atmospheric rendering. Expert ratings indicate a consistent, though not overwhelming, preference for ChatGPT in spatial composition, human-spatial responses, contextual fit, and strategic support for cognitive restoration, with a slight advantage for Gemini in visual realism. Across all cases, however, plan-to-image fidelity is limited, particularly for non-visual and operational patterns (e.g., sound, scent, thermal variability, circadian systems, infrastructure access). The findings suggest that current generative AI tools are best positioned as fast, co-creative aides for early exploration of biophilic ideas, rather than as reliable autonomous consultants for evidence-based, cognitively targeted biophilic design. Full article

Machine learning (ML) methods have been widely explored for predicting material properties. However, due to the rapid development of ML techniques and the diversity of available models, performance comparisons between traditional and graph-based machine learning models remain limited. Therefore, we evaluate 11 conventional ML models alongside the graph neural network-based Crystal Graph Convolutional Neural Network (CGCNN) for predicting three key properties—formation energy (E_f), band gap (E_g), and energy above hull (E_h)—across a dataset comprising single perovskites, double perovskites, and their combined structures. The results demonstrate that for single perovskites, CGCNN exhibits gains of over 20% in the root mean square error (RMSE) relative to the second-best model (Gradient Boosting Regression), achieving values of 0.205 eV/atom (E_f), 0.718 eV (E_g), and 0.167 eV/atom (E_h). Prediction accuracy for double perovskites is significantly enhanced by training CGCNN on a combined dataset, particularly for E_h, where the coefficient of determination (R²) improves approximately 68.1-fold compared to models trained exclusively on double-perovskite data. Feature importance analysis via one-shot, permutation-based, and recursive feature elimination (RFE) methods reveals that optimal model performance requires retention of at least the top 20 critical features. Furthermore, feature utilization patterns of CGCNN across different prediction tasks are visualized. This work provides actionable guidelines for model selection and feature engineering in perovskite property prediction, establishing a benchmark for future ML-driven materials discovery. Full article

Background: With the rapid advancement of artificial intelligence (AI), its applications in andrology are expanding across diagnostic assessment, preoperative planning, intraoperative assistance, and postoperative management. This narrative review aims to synthesize current evidence regarding AI applications across the spectrum of andrological surgery. Methods: A comprehensive literature search was conducted using the PubMed, Scopus and Web of Science databases to identify relevant studies published between January 2020 and October 2025. The search strategy utilized combinations of keywords including “artificial intelligence,” “andrology,” “erectile dysfunction,” “male infertility,” “microsurgery,” and “robotic-assisted surgery.” Original research and review articles published in English were selected based on their clinical relevance to surgical practice. Results: AI has shown promise in the evaluation and management of erectile dysfunction (ED), male infertility-related microsurgery, and complex reconstructive procedures. AI-based models can improve risk prediction and diagnosis of ED, standardize semen analysis, support individualized selection of surgical candidates for varicocele repair and other interventions, and augment microsurgery through enhanced visualization and decision support. In the postoperative phase, AI-driven tools are being explored for complication prediction, functional recovery monitoring, and long-term quality-of-life follow-up, enabling more patient-centered, continuous care. Conclusions: AI holds significant promise for advancing precision medicine in andrological surgery by enhancing objective assessment and intraoperative guidance. However, large-scale, standardized datasets and rigorous multi-institutional validation are needed. Establishing robust ethical and legal frameworks will be essential to ensure the safe and effective integration of AI into routine andrological care. Full article

Background/Objectives: Digital environments have intensified exposure to interpersonal stressors and appearance-related evaluation, raising concerns about cyberbullying and problematic internet use among women with eating disorders (EDs). This study examined whether cyberbullying exposure and problematic online use are associated with health-related quality of life in women receiving specialized outpatient care for eating disorders in Spain. Methods: We conducted a cross-sectional analysis of baseline data collected between 2018 and 2019 from a clinical cohort of 124 women in specialized ED treatment. ED symptoms were assessed using the SCOFF and the Bulimic Investigatory Test, Edinburgh (BITE). Problematic online use was measured with the Internet Addiction Test (IAT) and the Bergen Facebook Addiction Scale (BFAS), and cyberbullying exposure was summarized using a composite index. HRQoL was assessed with the EQ-5D index and visual analogue scale (EQ-VAS). Associations were examined using correlation analyses, group comparisons, and exploratory multiple linear regression models adjusting for age, body mass index (BMI), diagnosis, and comorbidity. Results: Cyberbullying exposure was strongly positively correlated with problematic internet and social media use (IAT and BFAS). Its bivariate associations with ED symptom measures were small and not statistically significant. Participants with physical or mental health comorbidities reported lower HRQoL on both the EQ-5D index and EQ-VAS scores (p < 0.01). In the exploratory adjusted regression model predicting EQ-5D, coefficients for cyberbullying exposure, IAT, and BITE severity were small and imprecisely estimated, whereas diagnosis category showed between-group differences (with the “other ED” category reporting lower EQ-5D scores relative to the reference group). The overall model explained approximately 26.7% of the variance in EQ-5D (adjusted R² = 0.22). Conclusions: In this clinical sample, digital-use measures co-occurred strongly with one another, and comorbidity was associated with poorer HRQoL at the bivariate level. In exploratory adjusted models, estimated associations of cyberbullying and problematic online use with HRQoL were imprecise, supporting cautious interpretation. Prospective and intervention studies are needed to determine whether digital interpersonal stressors contribute to HRQoL trajectories in women receiving specialized ED care and whether targeting these stressors improves patient-reported outcomes. Full article

Background: Poor sleep quality is common among university students and may contribute to adverse reproductive health outcomes, including dysmenorrhea. However, limited evidence exists on whether chronic sleep disturbance independently predicts dysmenorrhea severity or menstrual-related functional impairment after accounting for key confounders. Objectives: We aimed to determine the prevalence of poor sleep quality among female university students and to examine its association with (1) severe dysmenorrhea and (2) menstrual-related functional impairment. Methods: A cross-sectional study was conducted among female undergraduate students at Ras Al Khaimah Medical and Health Sciences University (United Arab Emirates). Sleep quality was measured using the Pittsburgh Sleep Quality Index (PSQI), and perceived stress was assessed using the Perceived Stress Scale (PSS-10). Dysmenorrhea severity was assessed using a 0–10 visual analog scale; functional impairment was defined as moderate/severe disruption in ≥1 life domain. Multivariable logistic regression models estimated adjusted odds ratios (aORs) for the association between sleep quality and menstrual outcomes after controlling for age, BMI, socioeconomic status, and stress. A component-level analysis examined independent effects of PSQI dimensions. Results: Of the 254 participants, 68.9% reported poor sleep quality and 48.8% reported severe dysmenorrhea. In adjusted models, moderate sleep problems (aOR = 2.00, 95% CI: 1.09–3.67, p = 0.024) and severe sleep problems (aOR = 3.63, 95% CI: 1.45–9.06, p = 0.006) were significantly associated with severe dysmenorrhea. Severe dysmenorrhea strongly predicted menstrual-related functional impairment (aOR = 4.81, 95% CI: 2.63–8.77, p < 0.001). Poor sleep quality remained independently associated with functional impairment (aOR = 1.96, 95% CI: 1.05–3.65, p = 0.035). In component analysis, sleep disturbance (PSQI Component 5) was the only independent predictor of severe dysmenorrhea (aOR = 2.11, 95% CI: 1.31–3.41, p = 0.002). Conclusions: Poor sleep quality, particularly sleep disturbance, is associated with increased odds of severe dysmenorrhea and menstrual-related functional impairment in female university students. Sleep fragmentation may represent a key mechanistic and modifiable contributor to menstrual pain severity. Integrating sleep assessment into dysmenorrhea management and evaluating sleep-focused interventions in longitudinal and interventional studies are warranted. Full article

Background and Clinical Significance: CEUS enhances the visualization of vascular patterns within liver lesions, enabling differentiation between benign and malignant lesions, including hemangiomas, focal nodular hyperplasia, and hepatocellular carcinoma, with high accuracy. Lymphangiomas are rare benign lymphatic-system tumors, with intra-abdominal lymphangiomas accounting for approximately 5% of cases, most of which occur in the pediatric population. Intra-abdominal lymphangiomas commonly occur in multiple localizations due to lymphangiomatosis, but solitary lymphangiomas in adults are rare and easy to be misdiagnosed due to asymptomatic cases or non-specific symptoms. Case Presentation: A 65-year-old male with a history of left nephroadrenalectomy due to clear renal-cell carcinoma and paraaortic lymphadenectomy (staging pT3bN0M0V1R0) presented for a routine contrast-enhanced abdominal computer tomography examination. The scan showed several hypervascular structures that accumulate contrast in the arterial phase in the right liver lobe. Three years later, the patient developed complaints of abdominal pain and night sweats. Multiple MRI and CT examinations were performed, followed by a CEUS and a liver-core biopsy, which supported the diagnosis of hepatic lymphangioma. Conclusions: CEUS may be a more valuable evaluation method for follow-up examination than repeating CT and MRI scans. The real-time diagnostic possibility and tissue-perfusion data provide more profound information about the lesion of interest. Thus, it can be used as a primary diagnostic tool when a biopsy is performed. Although this method is relatively new, it can be applied in clinical settings with great value, and it saves time and resources. Full article

To address the trade-off between parameter scale and generation quality in Vision-Language Models (VLMs), this study proposes a Multi-Feature Dynamic Instruction Tuning (MFDIT) image captioning model based on LLaMA. By integrating CLIP-based global features with SAM-derived local features, the model constructs a multi-level visual representation. Additionally, a Dynamic Prompt Adapter is designed to enable cross-modal semantic alignment with adaptive flexibility. Combined with a Low-Rank Adaptation (LoRA) fine-tuning strategy, the proposed method enhances the model’s capability in describing diverse images while training only 20 million parameters, accounting for merely 0.05% of the total parameter volume. Experimental results demonstrate that the model achieves a CIDEr score of 126.7 on the MSCOCO dataset, surpassing traditional adapter-based approaches by 3.0 points. Moreover, in the MME Benchmark evaluation, the proposed model outperforms the mainstream LLaMA-Adapter V2 by 7.3% and 3.8% in OCR and object counting tasks, respectively. Ablation studies further validate the synergistic effects of multi-feature fusion and dynamic instruction optimization. This research provides an efficient solution for parameter-efficient multimodal model training and potential deployment in resource-constrained environments. Full article

The Geological Strength Index (GSI) is a crucial tool for assessing jointed rock masses, but it is often hindered by subjectivity in visual assessments. In this study, we propose a novel quantitative GSI method wherein 3D laser-scanning point clouds are used to quantitatively derive empirical rock mass indices (SR and SCR) to estimate mechanical parameters. By integrating the GSI with the Rock Block Index (RBI) and joint spacing, a framework for quantifying the Structural Rating (SR) is established. Furthermore, the Analytic Hierarchy Process (AHP) is employed to assign weights to Surface Condition Rating (SCR) factors. The results indicate that infilling materials have the most significant impact on SCR (weight 0.6334), followed by weathering (0.2605) and roughness (0.1061). This method was applied to evaluate rock masses at depths of −915 to −960 m in the Sanshandao Gold Mine. The GSI values calculated for the foot wall, ore body, and hanging wall were 38.5, 33.8, and 37.8, respectively. Validation against conventional quantitative methods demonstrated high accuracy, with a maximum relative GSI difference of 1.5 and a deformation modulus difference of only 0.227 GPa. This data-driven approach effectively reduces subjectivity and provides a reliable tool for automated geotechnical parameter estimation. Full article

Sustainable urban development increasingly relies on hyperlocal environmental analytics created by smart city platforms that combine stationary and mobile sensors, Earth observations, meteorology, and land-use data. However, accurate spatio-temporal resolution can provide indirect identification and amplify cybersecurity threats. This article proposes the regulatory and technical mapping that implements the General Data Protection Regulation (GDPR) and the Network and Information Security Directive (NIS2) throughout the lifecycle of environmental data—reception, transport, storage, analytics, sharing, and publication. The methods combine doctrinal legal analysis, a review of the scope of recent research, formalized compliance modeling, modeling with synthetic city-scale datasets, expert identification, and demonstration of integrated analytics. The demonstration links deep evaluation of neural abnormalities (convolutional plus recurrent layers), short-term Fourier transformation of sensor signals, byte-to-image telemetry fingerprints, and protocol event counters, thereby tracking detection to explanatory evidence and to control actions. Deliverables include a matrix aligning lifecycle stages with GDPR principles and rights, as well as with the responsibilities of NIS2; a checklist for assessing the impact on data protection, which takes into account the risks of fairness and stigmatization; a basic set of controls for identification and access, secure design, monitoring, continuity, supplier assurance, and incident reporting; as well as a multi-layered publishing strategy that combines transparency with privacy through aggregation, delayed release, differentiated privacy budgets, and research enclaves. The visualization confirms that technical signals can be included in audit-ready reporting and automated response, while the guidelines legally clarify the relevant bases for common use cases such as air quality assurance networks, noise mapping, citizen sensor applications, and mobility and exposure modeling. The effects of the policy emphasize shared services for small municipalities, supply chain security, and ongoing review to counteract the mosaic effect. Overall, the study shows how cities can maximize environmental and social value based on environmental data, while maintaining privacy, sustainability, and equity by design. Full article

Objectives: To evaluate the effectiveness of zinc oxide (ZnO) ointments versus non-irritant barrier films (NIBFs) in the prevention of Incontinence-Associated Dermatitis (IAD). Specific objectives included analyzing the effects of treatment over time, establishing when IAD appears in each group, and determining the safety and cost-effectiveness of both treatments. Methodology: A multicenter prospective cohort study was carried out in 10 social health centers in Spain. The final sample included 164 older institutionalized patients with urinary and/or mixed incontinence, divided into two cohorts (79 with ZnO and 85 with NIBF). Follow-up lasted for six weeks. Validated scales were used for data collection, such as the Categorization of Moisture-Associated Skin Damage (MASD) of the GNEAUPP (National [Spain] group for the assessment of pressure ulcers and chronic wounds) and the Visual Erythema Scale (VES). Results: The overall incidence of IAD was 20.7% in the sample during follow-up. No statistically significant difference in effectiveness was found between ZnO (27.8% incidence of IAD) and NIBF (32.9% incidence of IAD) in preventing IAD (p = 0.479). However, survival analysis suggested that the onset of IAD is delayed more in the NIBF group. No adverse events or side effects were reported attributable solely to the use of the products. When considering the total cost per process (including staff application time), ZnO ointment was EUR 0.02 more expensive per patient per day than NIBFs. Conclusions: Although both products have similar efficacy and safety, NIBFs delay the onset of IAD more than ZnO. In addition, despite a higher unit price, NIBFs are more cost-effective per procedure because of savings in nursing time during application and removal. Full article

Acne severity grading is an important dermatological task that supports clinical diagnosis, treatment planning, and disease monitoring. Although self-supervised learning (SSL) has gained interest as a means to reduce reliance on large annotated datasets, its effectiveness for fine-grained and ordinal dermatological tasks remains unclear. This research systematically evaluates contrastive SSL for acne severity grading by comparing SimCLR-based pretraining with a diverse set of supervised deep learning models, including Convolutional Neural Networks and Vision Transformers, under controlled experimental conditions. The evaluation considers full-data training, label-scarce scenarios, and temperature tuning of the contrastive loss. The results consistently demonstrate the superiority of supervised transfer learning, which achieves Quadratic Weighted Kappa (QWK) scores ranging from 0.7616 to 0.8533. In contrast, SimCLR-based models exhibit substantially lower performance, with QWK values between 0.2343 and 0.4548 after fine-tuning. Although temperature adjustment achieved modest performance gains, it does not close this gap, with the best configuration attaining a QWK of 0.4548 using a ResNet18 backbone. Qualitative analysis using Grad-CAM further reveals that SimCLR-based contrastive SSL tends to exhibit diffuse attention patterns and limited localization of clinically relevant acne regions. Overall, these findings indicate that generic contrastive SSL objectives are poorly aligned with the subtle and localized visual cues required for acne severity grading. The results highlight the need for domain-aware representation learning in fine-grained dermatological image analysis. Full article

Generative Adversarial Networks (GANs) have emerged as a promising tool for augmenting medical image datasets used by AI solutions. However, GANs trained on small datasets (300–500 images) frequently encounter mode collapse, overfitting, and instability, which hinder their practical application. Many GAN-generated images look unrealistic. The Enhanced Deep Convolutional GAN (EDCGAN) is introduced to generate high-quality synthetic images of breast US (BUS). The model includes an experimental design for the Discriminator and Generator. The main components are spectral normalization (SN), the Squeeze-and-Excitation (SE) block, and the Scaled Exponential Linear Unit (SELU). One of the basic versions of DCGAN is considered for the proposed modifications. The stopping criteria are based on the convergence of the smoothed loss function and the constraints imposed on the Discriminator. The contribution is a combination of the above modifications and postprocessing based on the visual evaluation by radiologists and selected image processing metrics. The Inception Score (IS), the Structural Similarity Index (SSIM), and the Mean Squared Error (MSE) comply with the results obtained in the preceding works. The efficiency of augmenting the US data has been verified on a DL classification based on ResNet-18. The tests against training on a non-augmented data outperform ResNet by 5% and by the data augmented by the previous DCGAN by 3%. These numbers are substantial since this variant of ResNet has been pre-trained on 1000 categories by ImageNet-1K, including 1.28 million images. Additionally, the model wins the “Guess-the-real-image” game, competing with seven preceding GANs. Full article

The post-genomic era has transformed medical genetics, raising renewed debate over the role of medical cytogenetics in clinical practice. High-throughput sequencing and chromosomal microarray technologies now dominate cancer diagnostics, prenatal testing, and rare disease evaluation by enabling rapid detection of gene-level variation, often leading to the perception that cytogenetics is obsolete. However, this view overlooks the unique and complementary strengths of cytogenetic analysis. Although the relationship between cytogenetics and current NGS technologies can be compared to that between forests and trees versus leaves—both of which are necessary for clinical diagnosis—cytogenetic methods uniquely enable direct in situ visualization of chromosomes, allowing detection of large-scale structural and numerical genome alterations at the level of individual cells and cell populations. These system-level features that are frequently invisible or difficult to interpret using sequencing-based approaches alone yet are critical in disease contexts where genome architecture itself carries biological and clinical significance beyond individual genes. This article, therefore, advances a new perspective based on Genome Architecture Theory: that karyotype-level information organizes gene-level function and that many previous gene-centric genetic concepts require reexamination within a unified framework of clinical genomics. Rather than being replaced, cytogenetics is increasingly integrated with sequencing within a unified framework of clinical genomics that combines high-resolution molecular detail with system-level insight into genome organization. Reassessing the role of cytogenetics, therefore, has important implications for medical education, diagnostic strategy, and healthcare policy, as cytogenetics provides the appropriate platform for understanding system-level inheritance through karyotype coding and for advancing molecular medicine from a genome systems perspective. Full article