Search Results (683)

The ceramic defect detection technology based on deep learning suffers from the problems of scarce rare defect samples and class imbalance. However, the current deep generative image augmentation techniques are limited when applied to the task of augmenting rare ceramic defects due to issues such as uneven image brightness and insufficient features of small-sized defects, resulting in poor image quality and limited improvement in detection results. This paper proposes a ceramic rare defect image augmentation method based on TC-CycleGAN. TC-CycleGAN is based on the CycleGAN framework and optimizes the generator and discriminator structures to make them more suitable for ceramic defect features, thereby improving the quality of generated images. The generator is TC-UNet, which introduces the scSE and DehazeFormer modules on the basis of UNet, effectively enhancing the model’s ability to learn the subtle defect features on the ceramic surface; the discriminator is the TC-PatchGAN architecture, which replaces the original BatchNorm module with the ContraNorm module, effectively increasing the discriminator’s sensitivity to the representation of tiny ceramic defect features and enhancing the diversity of generated images. The image quality assessment experiments show that the method proposed in this paper significantly improves the quality of generated defective images. For the concave type images, the FID and KID values have decreased by 49% and 73%, respectively, while for the smoke stains type images, the FID and KID values have decreased by 57% and 63% respectively. The further defect detection experiments results show that when using the data set expanded by the method in this paper for training, the recognition accuracy of the detection model for rare defects has significantly improved. The detection accuracy of the concave and smoke stains types of defects has increased by 1.2% and 3.9% respectively. Full article

Reading proficiency in early childhood is crucial for academic success and intellectual development. However, more and more children are struggling with reading. According to the last PISA study in Austria, one out of five children is dealing with reading difficulties. The reasons for this are diverse, but an application that tracks children while reading aloud and guides them when they experience difficulties could offer meaningful help. Therefore, this proposal explores a prototyping approach for a core component that tracks children’s reading using a self-supervised Wav2Vec2 model with a limited amount of data. Self-supervised learning allows models to learn general representations from large amounts of unlabeled audio, which can then be fine-tuned on smaller, task-specific datasets, making it especially useful when labeled data is limited. Our model is operating on the phonetic level with the help of the International Phonetic Alphabet (IPA). To implement this, the KidsTALC dataset from the Leibniz University Hannover was used, which contains spontaneous speech recordings of German-speaking children. To enhance the training data and improve robustness, several data augmentation techniques were applied and evaluated, including pitch shifting, formant shifting, and speed variation. The models were trained using different data configurations to compare the effects of data variety and quality on recognition performance. The best model trained in this work achieved a phoneme error rate (PER) of 14.3% and a word error rate (WER) of 31.6% on unseen child speech data, demonstrating the potential of self-supervised models for such use cases. Full article

Goat meat represents a valuable source of high-quality protein and healthy lipids, although its consumption remains limited in Europe. This study aimed to evaluate the qualitative–quantitative changes in the nutritional, rheological, and sensorial characteristics of meat (Quadriceps femoris and Longissimus thoracis et lumborum muscles) from three different autochthonous goat breeds (Garganica, Derivata di Siria, and Capra di Potenza) and a cosmopolitan, genetically selected one (Saanen), reared in Basilicata (Italy), during a 7-day wet aging process. Forty kids (10 per breed) were slaughtered at 50 ± 3 days, and meat samples were vacuum-aged at 4 ± 1 °C and analyzed at 0, 3, and 7 days. Data showed that genotype was the main factor influencing meat quality, while wet aging mainly improved rheological parameters, particularly in LTL muscles. Notably, Capra di Potenza exhibited the most favorable fatty acid profile, with lower atherogenic (average values of 0.80 in LTL and 0.92 in QF) and thrombogenic (average values of 1.49 in LTL and 1.59 in QF) indices, whereas Derivata di Siria showed the greatest oxidative stability (average values of 0.060 in LTL and 0.036 in QF). Overall, local breeds of kids’ groups produced more tender and aromatic meat than Saanen. These findings highlight the potential of native goat breeds for premium meat production and suggest an effective post-mortem aging technique to enhance their quality, promoting the diffusion of niche products as well as biodiversity preservation. Full article

Background/Objectives: Suicidal behaviors are a major clinical concern in adolescents, particularly among those with disorders marked by emotion dysregulation and impulsivity. Although attention-deficit/hyperactivity disorder (ADHD) and borderline personality disorder (BPD) each heighten suicide risk, little is known about whether their occurrence confers additive or interactive effects in youth. This study examined whether ADHD and BPD diagnoses show additive or interactive associations with the suicide risk in adolescents. Methods: In this cross-sectional observational clinical study, the sample included 108 Polish adolescents (66.7% female; aged 13–17 years) recruited from inpatient and outpatient psychiatric settings (Independent Public Healthcare Facility, Children and Youth Treatment Center in Zabór, the Youth Sociotherapy Center No. 2 in Wrocław, and the District Educational Center in Jerzmanice-Zdrój (Poland)). The data collection for our study was conducted between May 2024 and July 2025. Diagnoses and suicide risk were assessed using the Mini International Neuropsychiatric Interview for Children and Adolescents (MINI-KID 7.02). Associations of ADHD and BPD with suicide risk were tested using linear and logistic regression models while accounting for age, sex, the current depressive episode, and the use of psychiatric medications. Results: Unadjusted analyses revealed significant main, but not interactive, associations of BPD and ADHD with suicide risk. When covariates were included in the model, BPD remained strongly associated with suicidality severity and with the presence of any suicide risk (adjusted OR = 7.00, 95% CI [1.55–31.57]), whereas the association between ADHD and suicidality was attenuated and did not reach conventional levels of statistical significance (adjusted OR = 3.48, 95% CI [0.93–13.08]). No statistically detectable ADHD × BPD interaction was observed. Estimates for ADHD were directionally consistent across models but characterized by wide confidence intervals. Conclusions: Adolescents with BPD appear to be at particularly high risk of suicide and should receive focused assessment, safety planning, and early intervention as part of routine care. In contrast, suicidality among adolescents with ADHD appears to be influenced by co-occurring clinical conditions, and its independent association with suicide risk remains statistically uncertain after adjustment. Clinicians should therefore remain alert to suicidality in youth with ADHD, while paying particular attention to accompanying symptoms and comorbid diagnoses that may further increase risk. Full article

Unlike high-dose scans, low-dose cardiac CT perfusion imaging reduces patient radiation exposure and thereby the risk of potential health effects. However, it introduces significant image noise, degrading diagnostic quality and limiting clinical assessment. Denoising is thus a critical preprocessing step to enhance image quality without compromising anatomical or perfusion details. Traditionally used reconstruction-domain methods, such as Iterative Reconstruction and Compressed Sensing, are often limited by algorithmic complexity, dependence on raw sinogram data, and restricted adaptability. Conversely, image-domain methods offer more adaptable denoising options. Recently, learning-based approaches have further expanded this flexibility and demonstrated state-of-the-art performance across various denoising tasks. In this work, we present a deep learning-based denoising method specifically tuned for low-dose cardiac CT perfusion imaging. Our model is trained to reduce noise while preserving structural integrity and temporal contrast dynamics, which are critical for downstream analysis. Unlike many existing methods, our approach is optimized for perfusion data, where temporal consistency is essential. Residual cardiac motion remains a separate challenge, which we aim to address in our future work. Experimental results show significant improvements in quantitative image quality, using both reference-based and no-reference metrics, such as MSE/PSNR/SSIM and NIQE/FID/KID, as well as improved accuracy of perfusion measurements. Full article

Goat meat is widely valued as a healthy option due to its lean nature, yet strategies to further optimize its intrinsic nutritional composition remain a key objective. This study examined the influence of melatonin on muscle development and visceral fat deposition in cashmere goats, focusing on its role in augmenting systemic antioxidant capacity and modifying gut microbiota. Thirty goat kids were randomly assigned to a control or a melatonin-treated (2 mg/kg body weight) group. Melatonin implantation induced a metabolic shift characterized by reduced visceral fat deposition (perirenal, omental, and mesenteric fat; p < 0.05) without impacting intramuscular fat. Concurrently, it promoted muscle accretion, as demonstrated by an increase in crude protein content and hypertrophy of muscle fibers in the Longissimus thoracis et lumborum, Gluteus medius, and Biceps femoris muscles (p < 0.05). These effects were underpinned by an enhanced systemic antioxidant capacity (elevated CAT, GSH-Px, T-AOC, and reduced MDA; p < 0.05), changes in gut microbiota, and a concomitant improvement in gastrointestinal morphology, evidenced by increased rumen papilla length and intestinal villus height. Melatonin enriched beneficial genera (e.g., Succiniclasticum, Butyrivibrio, Akkermansia), which were significantly correlated with reduced adiposity and improved protein deposition. These improvements resulted from the concerted actions of an enhanced systemic antioxidant defense system and a beneficially modulated gut microbial community. This trial observed no effect on intramuscular fat deposition, suggesting that improving intramuscular fat may require a systematic fattening regimen. This study provides a scientific foundation for employing melatonin as a nutritional strategy in goat production to improve meat quality. Full article

Background/Objectives: Keratitis–ichthyosis–deafness (KID) syndrome is an exceptionally rare congenital multisystem disorder, with an estimated prevalence below 1:1,000,000 and fewer than 100 reported cases worldwide. It is characterized by hyperkeratosis, alopecia, nail dystrophy, hearing loss, and ocular involvement. While dermatological management is well described, surgical experience with wound treatment in KID syndrome remains extremely limited. The objective of this report is to describe the surgical management and outcomes of a patient with chronic lower-limb wounds associated with KID syndrome. Methods: A 35-year-old male with KID syndrome was referred to the surgical outpatient clinic for chronic traumatic wounds of both lower limbs resistant to conservative dermatological therapy. Initial outpatient treatment included serial sharp surgical debridement under local anesthesia, combined with topical keratolytics and silver sulfadiazine with hyaluronic acid. Due to minimal improvement after three months, an inpatient surgical approach was initiated, involving complete excision of hyperkeratotic tissue, creation of a wide ulcer bed, and adjunct systemic and topical therapies. Results: The inpatient management enabled thorough removal of pathological tissue and better control of local infection and inflammation. Combined systemic antibiotic and antipsoriatic therapy, together with topical sodium hyaluronate and iodine, was associated with gradual wound healing and improved skin condition. The patient tolerated the procedures well, without major complications. Conclusions: Surgical debridement and excision, when combined with targeted dermatological and antimicrobial therapy, can be an effective and safe strategy for managing chronic wounds in KID syndrome. This case highlights the importance of multidisciplinary collaboration and individualized surgical planning in this extremely rare disorder. Full article

The specific objective of the current paper was the description of some health-related outcomes in sheep and goat farms in Greece with common grazing with wildlife ruminants. Faecal samples were collected and information regarding health management applied on farms was obtained from sheep and goat farms located throughout Greece (325 and 119 farms, respectively). Common grazing of livestock (sheep, goats) with wildlife ruminants (roe deer, red deer) was reported in 9.2% of farms. Faecal epg counts in farms with common grazing with wildlife ruminants were higher than in farms with no common grazing: respective median values were 270 epg versus 150 epg; also, there was a tendency for faecal counts >300 epg to be recorded more frequently among the former farms (36.1% versus 23.1% of farms). Median annual incidence of cases of abortion and of cases of diarrhoea in lambs/kids was significantly higher in farms with common grazing with wildlife ruminants: 1.7% and 9.0%, respectively, versus 0.0% and 1.7%, respectively, in farms with no common grazing. The findings have revealed associations of common grazing with wildlife ruminants with the health of sheep and goats. These can have implications in the health management of farms, for example, in the administration of anthelmintic treatments and in the development of vaccination programmes in livestock farms. Full article

Microscopic wood images are vital in wood analysis and classification research. However, the high cost of acquiring microscopic images and the limitations of experimental conditions have led to a severe problem of insufficient sample data, which significantly restricts the training performance and generalization ability of deep learning models. This study first used basic image processing techniques to perform preliminary augmentation of the original dataset. The augmented data were then input into five GAN models, BGAN, DCGAN, WGAN-GP, LSGAN, and StyleGAN2, for training. The quality and model performance of the generated images were assessed by analyzing the degree of fidelity of cellular structure (e.g., earlywood, latewood, and wood rays), image clarity, and diversity of the images for each model-generated image, as well as by using KID, IS, and SSIM. The results showed that images generated by BGAN and WGAN-GP exhibited high quality, with lower KID values and higher IS values, and the generated images were visually close to real images. In contrast, the DCGAN, LSGAN, and StyleGAN2 models experienced mode collapse during training, resulting in lower image clarity and diversity compared to the other models. Through a comparative analysis of different GAN models, this study demonstrates the feasibility and effectiveness of Generative Adversarial Networks in the domain of small-sample image data augmentation, providing an important reference for further research in the field of wood identification. Full article

Objective: This study aimed at detecting changes in redox balance (reactive oxygen species (ROS), serum antioxidant capacity (SAC) and oxidative stress index (OSi)) during the peripartum period in dairy goats and evaluating the relationship between oxidative biomarkers (ROS, SAC and OSi) and selected metabolic biomarkers (total cholesterol, triglycerides, beta-hydroxybutyrate (BHB) and non-esterified fatty acids (NEFAs)). Materials and Methods: Blood samples were collected from 32 secondiparous Alpine dairy goats (average daily milk production: 3.93 ± 1.23 L) from the same commercial herd at dry-off, kidding and 30 days in milk (DIM). Results: Fatty acids at dry-off and BHB at kidding (rho: 0.56; p < 0.01) and 30 DIM (rho: 0.54; p < 0.01) were positively correlated, suggesting a potential role of pre-partum NEFA concentrations on the metabolic status of dairy goats during early lactation. Parturition was associated with the highest values of ROS (183.13 ± 6.99 Carr. U; p < 0.05) and OSi (0.50 ± 0.03 Carr. U/umol HCLO/mL; p < 0.05) probably due to the stress typical of this period. As expected, OSi exhibited a positive correlation with ROS (rho: 0.405, p < 0.01) and a negative correlation with SAC (rho: −0.707, p < 0.01). Furthermore, NEFAs showed a tendency to be positively correlated with ROS (rho: 0.191 p < 0.06) and were positively correlated with OSi (rho: 0.219 p < 0.03), suggesting a potential role of this metabolic parameter on oxidant status. Conclusions: The knowledge of the interplay between oxidative stress and metabolic changes during the peripartum period could potentially facilitate the development of strategies for the early detection and management of metabolic disorders in dairy goats. Full article

Generative adversarial networks (GANs) typically require large datasets for effective training, which poses challenges for volumetric medical imaging tasks where data are scarce. This study addresses this limitation by extending adaptive discriminator augmentation (ADA) for three-dimensional (3D) StyleGAN2 to improve generative performance on limited volumetric data. The proposed 3D StyleGAN2-ADA redefines all 2D operations for volumetric processing and incorporates the full set of original augmentation techniques. Experiments are conducted on the NoduleMNIST3D dataset of lung CT scans containing 590 voxel-based samples across two classes. Two augmentation pipelines are evaluated—one using color-based transformations and another employing a comprehensive set of 3D augmentations including geometric, filtering, and corruption augmentations. Performance is compared against the same network and dataset without any augmentations at all by assessing generation quality with Kernel Inception Distance (KID) and 3D Structural Similarity Index Measure (SSIM). Results show that volumetric ADA substantially improves training stability and reduces the risk of a mode collapse, even under severe data constraints. A strong augmentation strategy improves the realism of generated 3D samples and better preserves anatomical structures relative to those without data augmentation. These findings demonstrate that adaptive 3D augmentations effectively enable high-quality synthetic medical image generation from extremely limited volumetric datasets. The source code and the weights of the networks are available in the GitHub repository. Full article

Background/Objectives: There is a lack of food frequency questionnaires (FFQs) specifically designed to assess food consumption based on processing levels in younger populations. This study evaluates the validity and the reliability of a 107-item FFQ (NFFQ-Kids) in estimating the intake (g/day) and the weight and energy ratios of Nova groups in Italian children and adolescents aged 2–18 years. Methods: The NFFQ-Kids was administered twice (T0 and T1), with a four-week interval. A total of 73 participants completed the questionnaire at T0, and 53 completed it at T1. Participants were selected from the ICARO Study (December 2023–April 2024), a web-based cohort study in Southern Italy. Questionnaire validity was assessed by comparison with a 3-day food record (two weekdays and one weekend day) collected between the two NFFQ-Kids administrations. Test–retest reliability was evaluated to assess the consistency of the questionnaire over time. Results: A moderate correlation was found between the NFFQ-Kids and the 3-day food record for the energy ratio of ultra-processed foods (UPFs) (r = 0.48; p < 0.001; ICC = 0.46; 95% CI 0.29–0.64) and the weight ratio for both unprocessed or minimally processed foods (r = 0.49; p < 0.001; ICC = 0.50; 95%CI 0.33–0.66) and ultra-processed foods (UPFs) (r = 0.58; p < 0.001; ICC = 0.58; 95%CI 0.42–0.72). Overall, the NFFQ-Kids showed good test–retest reliability across all Nova group intakes, total food, and relative energy and weight ratios of UPFs (r = 0.71, ICC = 0.67; and r = 0.79, ICC = 0.67, respectively), indicating stable measurements over time. Conclusions: The NFFQ-Kids demonstrated acceptable validity and good reliability, proving useful for assessing food intake by processing level in Italian youth. Full article

Background/Objectives: Chronic inflammatory diseases (CIDs) often present a preclinical phase influenced by genetic and environmental factors, including nutrition. Early dietary habits may modulate long-term health trajectories by shaping the intestinal microbiota. Previous work showed that weaning with fresh foods from the Mediterranean diet (MD) improved dietary habits and microbiota composition at 3 years of age. This study aimed to assess whether such benefits persist at 9 years. Methods: This long-term follow-up included 191 children (96 MD, 95 controls) from the original randomized cohort (ClinicalTrials.gov NCT05297357). The primary endpoint was adherence to MD (KidMed score ≥ 8). Secondary endpoints included BMI, incidence of CID, maternal dietary adherence, and intestinal microbiota composition in a subset of 36 children. Results: At 9 years, no difference was found in overall MD adherence (27.4% controls vs. 27.1% MD; p > 0.99) or BMI (17.7 vs. 18.1 kg/m²; p = 0.384). However, children from the MD group reported higher daily vegetable intake (71.9% vs. 51.6%; p = 0.005). Microbiota analyses revealed persistent differences between groups, with higher alpha diversity in the MD group. Although not statistically significant, the MD group showed lower prevalence of atopic dermatitis, allergic rhinitis, autism spectrum disorder, and ADHD. Maternal adherence to MD did not differ between groups. Conclusions: Early introduction of Mediterranean-style foods during weaning exerts lasting effects on dietary patterns and gut microbiota, with a potential protective trend against CID. While overall MD adherence converged between groups by 9 years, these findings suggest that early-life nutritional interventions may induce durable microbiome-mediated benefits and contribute to preventive strategies for chronic disease, warranting confirmation in larger, extended cohorts. Moreover, this study highlights the value of the collaboration between the Italian primary care pediatric system and the integration of the pediatric residency program, demonstrating a feasible and cost-effective methodology to generate large-scale prospective data within routine clinical practice. Larger studies and a longer follow-up period are necessary to confirm these results. Full article

This study aimed to investigate the effects of rumen-protected methionine (RPM) on the production performance of lactating dairy goats. Thirty first-time lactating Guanzhong dairy goats with identical kidding dates and comparable body weights (41.17 ± 3.05 kg) were randomly assigned to two groups: (1) CON: basal diet and (2) RPM: basal diet + 7.5 g/day RPM. The duration of the experiment was 21 days. Compared with the CON group, the RPM group presented a significant increase in milk yield, 4% fat-corrected milk (FCM), and feed efficiency; however, no significant difference was observed in dry feed intake (DMI). Moreover, milk fat, protein, lactose, and SNF production was greater in the PRM group than in the CON group. Compared with the CON group, the RPM group presented higher nonesterified fatty acid (NEFA) and very-low-density lipoprotein (VLDL) levels, and no significant differences in the other metabolites were detected. The concentrations of acetate, propionate, and total volatile fatty acids (TVFAs) in the feces of the RPM group were significantly greater than those in the CON group; however, no significant differences were detected in the concentrations of isobutyrate, butyrate, and valerate. Furthermore, genera such as Muribaculaceae, Bifidobacterium, and Christensenellaceae were significantly enriched in the feces of the RPM group. Concurrently, Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis revealed that the RPM group was significantly enriched in pathways associated with amino acid synthesis, the immune system, and energy metabolism. In summary, dietary supplementation with RPM improved the lipid metabolism function of the liver, increased the abundance of beneficial bacteria such as Muribaculaceae and Bifidobacterium in the colon, and enriched microbial functions related to energy and amino acid metabolism, thereby enhancing colon fermentation and host metabolic status, ultimately improving the production performance of lactating dairy goats. These findings elucidate the positive effects of RPM on the production performance and metabolic health of dairy goats, potentially offering new perspectives and strategies for optimizing dairy production. Full article

Acute Kidney Injury (AKI) following endovascular aortic repair (EVAR) is often diagnosed too late using conventional markers, limiting opportunities for timely intervention in this high-risk population. We investigated whether a mechanism-based biomarker panel could provide improved early AKI detection in EVAR patients. This prospective, single-center study enrolled 68 consecutive EVAR patients between April 2022 and June 2024. AKI was diagnosed using KDIGO 2012 criteria. Seven novel biomarkers, including Proenkephalin A 119-159 (penKid), Semaphorin-3A (SEMA-3A), Retinol Binding Protein-4 (RBP-4), Kidney Injury Molecule-1 (KIM-1), Netrin-1, Tissue Inhibitor of Metalloproteinases-2, and Insulin-Like Growth Factor Binding Protein-7, were measured at baseline, immediate postoperative, 24 h, and 48 h time points, and selected based on distinct nephron locations and release mechanisms. AKI occurred in 18 (26.5%) patients. Top-performing individual biomarkers included serum SEMA-3A (AUC 0.88), serum RBP-4 (AUC 0.81), and penKid (AUC 0.76). A three-biomarker panel combining serum penKid, serum SEMA-3A, and urinary KIM-1 achieved robust discriminatory performance (AUC 0.89, 95% CI 0.77–1.00), superior to individual biomarkers. An alternative panel with serum RBP-4 demonstrated comparable performance (AUC 0.81, 95% CI 0.65–0.99). Multi-biomarker panels combining functional, stress, and injury markers demonstrate promising performance for early AKI detection in EVAR patients. External validation in independent, multi-center cohorts is required before clinical implementation. Full article