Cd, Cr, and Pb concentrations were analysed in 18 bottled drinking water samples from the Bosnia and Herzegovina (B&H) market, classified as natural spring water or still natural mineral water. Metal concentrations were determined using atomic absorption spectrometry with electrothermal atomisation in a graphite furnace. Cr was quantified in two samples, Pb in four samples, while Cd concentrations were below the limit of detection in all samples. All concentrations were below national regulatory limits. Health risks associated with Cr and Pb exposure through bottled water consumption were assessed for adolescents, adults, and pregnant women by calculating the hazard quotient (HQ), hazard index (HI), carcinogenic risk (CR), and total carcinogenic risk (TCR). Some exposure scenarios exceeded HQ and HI thresholds, suggesting potential non-carcinogenic effects, primarily due to Pb exposure. CR and TCR values across all exposure scenarios were below established thresholds, indicating generally acceptable or negligible carcinogenic risk. Given the limited sample set, the data on Cd, Cr, and Pb content, along with the health risk assessment, should be regarded as preliminary, serving as a starting point to guide future research. Considering the health implications of heavy metal exposure, continuous water-quality monitoring and a more comprehensive health risk assessment, including additional heavy metals and a larger sample set, are essential, as regulatory compliance does not necessarily imply the absence of health risks.

Dementia is a progressive condition that impairs cognitive processes such as memory, decision making, and the ability to manage daily activities. Recent estimates suggest that more than half of all dementia cases could be preventable by addressing their risk factors, including disease comorbidities such as diabetes and vision loss. Yet, we lack a comprehensive molecular map of dementia comorbidities. In this work, we analyzed Austrian nationwide hospital claims data, comprising 13 million hospital stays from 2015 to 2019, to systematically assess dementia-related risk across disease comorbidity patterns, covering both their molecular relationships and their epidemiological overrepresentation. We identified disease trajectories occurring before and at the time of dementia diagnosis, revealing both sex-specific and shared comorbidity patterns. Overall, we identified 51 potential risk factors, with a prominent contribution from endocrine and metabolic disorders. While Parkinson's disease emerged as a strong molecularly related driver of dementia, we also identified emerging and previously under chracterized risk factors, including vitamin D deficiency. This integrative framework provides a comprehensive view of dementia associated disease networks and identifies novel, potentially modifiable risk factors. These results offer new opportunities for targeted prevention strategies and advance our understanding of the complex interplay between comorbidities and dementia development.

Tumor evolution is driven by various mutational processes, ranging from single-nucleotide variants (SNVs) to large structural variants (SVs) to dynamic shifts in DNA methylation. Current short-read sequencing methods struggle to accurately capture the full spectrum of these genomic and epigenomic alterations due to inherent technical limitations. To overcome that, here we introduce an approach to identify and analyze the genomic and epigenetic events in different stages of tumoral evolution from long-read sequencing of single-cell derived sublines. We then use it to profile 23 sublines of a mouse cutaneous melanoma cell line, characterized with distinct growth phenotypes and treatment responses. We develop a computational framework for harmonization and joint analysis of different variant types in the evolutionary context. Uniquely, our framework enables detection of recurrent amplifications of putative driver genes, generated by independent SVs across different lineages, suggesting parallel evolution. In addition, our approach revealed gradual and lineage-specific methylation changes associated with aggressive clonal phenotypes. We also show our set of phylogeny-constrained variant calls along with openly released sequencing data can be a valuable resource for the development and benchmarking of computational methods. Tumor evolution involves genetic and epigenetic changes that are difficult to resolve with standard sequencing approaches. Here, authors use long-read sequencing of single cell-derived melanoma sublines to map mutations, structural variants and DNA methylation, revealing parallel genomic changes and lineage-specific epigenetic trajectories linked to tumor behavior.

Background: Plavac mali is a well-known native Croatian cultivar used for the production of premium wines. One of the locations known for this variety is the island of Hvar, in the southern Dalmatia subregion. Grape quality is influenced by the canopy microclimate, and one of the main practices for manipulating the canopy is defoliation. Aim: The aim was to investigate the impact of defoliation treatments on grape composition in two vineyards with different microclimates. Materials and Methods: The study was carried out in 2020 at two sites on the island of Hvar: the Selca vineyard, situated inland, and Sveta Nedjelja, situated on the southern side of the island. The experiment was set up as a randomized block design with three replicates of 20 vines per treatment. The defoliation treatments were control (C), basal defoliation before véraison (T1), basal defoliation at véraison (T2), and apical defoliation above the fruiting zone at véraison (T3). Total soluble solids (TSS), titratable acidity, and pH were determined according to OIV (2019) methods, and organic acids (tartaric, malic, citric) by HPLC. Individual polyphenolic compounds were extracted from freeze-dried grape skins and quantified by HPLC with diode-array and fluorescence detection, with results expressed as mg/kg of skin dry weight. Data were analyzed by one-way ANOVA with Duncan's multiple range test (p < 0.05) in XLSTAT. Results and Discussion: At both sites, the defoliation treatments affected TSS and polyphenolic profiles, increasing TSS and reducing the content of polyphenols. Treatments at Sveta Nedjelja accumulated excessive TSS, reaching over 110 °Oe. The highest content of polyphenolic compounds was found in T2 at the Selca vineyard and in T1 at the Sveta Nedjelja vineyard. Conclusion: In the context of a warm Mediterranean climate, this viticultural practice and its performance should be reconsidered.

Introduction: This study evaluated the accuracy of various cone-beam computed tomography (CBCT) reconstruction modalities, specifically the HyperSight (HS) detector, in comparison to standard computed tomography (CT) simulation for potential use in online adaptive radiotherapy. The research focused on the Hounsfield Unit (HU) to relative electron density (RED) conversion and its subsequent impact on volumetric modulated arc therapy (VMAT) dose calculations. Methods: Two tissue-equivalent phantoms, the advanced electron density and the CIRS Thorax phantom, were utilized. Imaging was performed on a SOMATOM go.Open Pro CT simulator and a Varian TrueBeam medical linear accelerator using CBCT, HS-CBCT, HS-iterative CBCT (iCBCT) and HS-iCBCT metal artifact reduction protocols. Calibration curves (HU-RED) were generated for two regions: RED <1.2 and 1.2≤ RED <1.8. VMAT plans (6 MV) were created in the Eclipse 18.1 treatment planning system (TPS), using an anisotropic analytical algorithm (AAA) and Acuros XB algorithm. Absolute dose measurements were conducted using an SNC125c ionization chamber and compared with TPS-calculated doses. Results: In the soft-tissue region (RED <1.2), all imaging modalities showed an excellent linear correlation with CT (r > 0.998), with HU deviations within ± 30 HU. In the high-density region (1.2 ≤ RED < 1.8), HS-CBCT demonstrated superior stability with the lowest root mean square error, 62.88 HU. Dosimetric results showed that 96.7% of all measurement points met the ± 3–4% agreement criteria. For the AAA algorithm, HS-iCBCT exhibited the highest precision (standard deviation = 0.59) and the lowest mean absolute error. The Friedman test confirmed a statistically significant difference between modalities (p < 0.05), with HS-iCBCT showing the most consistent performance. Conclusion: Both HS-CBCT and HS-iCBCT provide highly accurate HU-RED conversions and reliable dosimetric results for RED < 1.8.

OBJECTIVE To assess whether transfer-learning models applied to panoramic radiographs (PANs) can classify individuals at the threshold of legal majority (≥18 years). MATERIALS AND METHODS Vision Transformer (ViT) and EfficientNetV2 models were trained on PANs from Bosnian and Lebanese individuals aged 14-24.99 years (n = 1764), considering pooled and sex-specific datasets with and without augmentation. Binary classification, multiclass classification, and regression models were trained and evaluated. Model performance on an internal test set derived from the same sample was summarized using accuracy, sensitivity, specificity, F1 score, and area under the receiver operating characteristic curve (ROC AUC). For regression and multiclass models, legal majority classification was additionally assessed by thresholding predicted ages or age categories at 18 years. External validation employed an independent Brazilian dataset (n = 1579; 14-24.99 years). Formal statistical comparison between internal and external performance employed two-proportion z-tests for accuracy and DeLong's test for ROC AUC. RESULTS On the internal test set, the best-performing binary classification model, EfficientNetV2 with augmentation on the pooled dataset, achieved an accuracy of 0.90, sensitivity of 0.92, specificity of 0.86, F1 score of 0.91, and ROC AUC of 0.93. Using the same pooled, augmented configuration, thresholded regression predictions achieved accuracies of 0.85 for EfficientNetV2 and 0.83 for ViT, whereas thresholded multiclass predictions achieved accuracies of 0.84 and 0.73, respectively. Compared with direct binary classification, these thresholded outputs showed no clear advantage, and multiclass models generally showed higher specificity but lower sensitivity. This same pattern was retained when thresholded models were evaluated on the external validation set, with thresholded regression remaining comparatively stable and thresholded multiclass performance declining more markedly, especially for ViT. Visualization maps (gradient-weighted class activation mapping [Grad-CAM] and occlusion sensitivity) confirmed attention to relevant dental structures. On external validation, the same binary EfficientNetV2 configuration achieved an accuracy of 0.81, sensitivity of 0.90, specificity of 0.61, F1 score of 0.87, and ROC AUC of 0.85. Sex-specific models performed similarly, showing no clear advantage over pooled training. Statistical testing confirmed significant AUC degradation on external validation for both models (EfficientNetV2: p = 0.004; ViT: p < 0.001), while the accuracy drop was significant only for ViT (p = 0.035). CONCLUSIONS This study demonstrated that transfer learning with EfficientNetV2 and Vision Transformer can distinguish minors from adults using PANs with high internal performance and acceptable external generalization. Direct binary classification provided the most robust approach for legal age assessment in the present dataset. The findings support the forensic potential of these models, while also indicating the need for further work to improve robustness and real-world applicability.

BackgroundThe comet assay is a sensitive and widely used technique for assessing DNA damage at the single-cell level. Despite its advantages, traditional manual scoring methods remain time-consuming, subjective and limited in scalability, posing challenges for high-throughput and standardized analysis.ObjectiveThis study aims to develop and evaluate a deep learning-based system for automated comet assay image classification, addressing limitations of manual and semi-automated approaches while enhancing accuracy, reproducibility and processing efficiency.MethodA YOLOv5-based object detection model was trained on a dataset of 875 annotated comet assay images, curated through a three-step expert-reviewed process. Various hyperparameters and data augmentation techniques were optimized to improve performance. The dataset was split into training, validation and test sets, and model performance was evaluated using mAP, precision, recall and confusion matrix analysis.ResultsThe model achieved strong performance, with mAP@0.5 reaching 0.98 and recall exceeding 0.8. Detailed analyses revealed robust learning behavior and generalization capacity. Visual outputs, including precision-recall curves and class-wise confusion matrices, confirmed high classification accuracy, although overlapping comet structures and class imbalance posed challenges. The model demonstrated improved scalability and processing speed compared to traditional tools, supporting its integration into web-based applications.ConclusionThe proposed YOLOv5-based system offers a scalable and accurate solution for automating comet assay analysis. It significantly enhances throughput and reduces human error, supporting its application in genotoxicity testing, biomonitoring and molecular epidemiology. Future work will focus on handling overlapping structures, benchmarking against existing tools and optimizing deployment in real-world laboratory settings.

The calcium (Ca2+) sensor calmodulin (CaM) genes CALM1, CALM2, and CALM3 were recently included in the American College Medical Genetics and Genomics (ACMG) secondary findings (SF) list, given their significance in causing long QT syndrome (LQTS) and catecholaminergic polymorphic ventricular tachycardia (CPVT). These three genes share identical protein sequences, posing potential challenges in variant interpretation. Using paralogue annotation (PA) to classify pathogenic variants and variants of uncertain significance (VUS) across these three paralogue genes, we performed a systematic, semi-automated curation of the CALM1, CALM2, and CALM3 variants. The analysis identified 173 unique CALM variants from ClinVar and Mastermind databases (75 CALM1, 59 CALM2, and 39 CALM3 variants). After paralogue annotation, we identified 126 unique variants in each of the three genes—378 cDNA variants in total. Out of 126 unique variants for each CALM gene, 63 were VUS, 62 were likely pathogenic/pathogenic (LP/P), and one was conflicting (192 VUS, 186 P/LP, and 3 C in total). Twelve unique variants in the CALM1, CALM2, or CALM3 genes had conflicting classifications between VUS and LP/P calls, which were resolved as LP/P. The application of paralogue annotation and variant curation resulted in an increased number of likely pathogenic/pathogenic variants (111% increase). Additionally, our analysis confirms that the majority of known pathogenic variants are predominantly located within the C-lobe of the CaM protein. This study highlights the benefits of paralogue annotation for accurate variant interpretation in CALM1, CALM2, and CALM3 genes, suggesting that a reduction in missed diagnoses is associated with calmodulinopathies.

The AMY2B gene encodes pancreatic amylase, a critical enzyme for starch digestion. While previous studies have examined AMY2B copy number variation (CNV) in domestic and some wild animals, less is known about wild carnivores inhabiting regions with limited anthropogenic starch exposure. We analyzed blood samples for serum amylase activity and copy number variation in AMY2B gene from 8 wolves (Canis lupus), 11 brown bears (Ursus arctos), and 3 red foxes (Vulpes vulpes) from Bosnia and Herzegovina. AMY2B gene copy number was assessed using droplet digital PCR (ddPCR), and serum amylase activity and glucose levels were quantified. Although the number of fox samples was limited, foxes and wolves consistently harbored two copies of AMY2B, while brown bears exhibited higher CNV (3.67–8.40, mean 5.88). Serum amylase activity was highest in foxes, moderate in wolves, and variable but lower in bears. Despite differences in AMY2B copy number and serum amylase activity, circulating glucose concentrations did not differ significantly among species. Our findings suggest that variation in AMY2B copy number among wild carnivores may be associated with species-specific evolutionary histories and dietary adaptations, providing insight into genomic mechanisms underlying carbohydrate utilization in natural populations.

The purpose of this paper is to develop machine learning (ML) models for prediction of surface roughness and cutting forces of 42CrMo4 steel in hard turning process. A full factorial experimental design with four input parameters: cutting speed, depth of cut, feed and insert radius was used to develop ML models for predicting the performance of turning process. The backward linear regression, random forest (RF) and XGBoost were used. Also, for the linear regression model and for the best RF and XGBoost model five-fold cross validation was done to confirm that the models provide reliable generalization estimates rather than performance dependent on a single data split. The XGBoost model demonstrates the most compact clustering of residuals with fewer large errors, indicating better overall stability and predictive consistency compared to the linear regression and RF models. The application of different ML methods with monitoring of standardized residuals on unseen data confirms the reliability of the developed models in real application conditions. This study provides a structured and comparative modeling framework across multiple output variables, where backward linear regression, RF and XGBoost models were developed. Several architectural and hyperparameter variations of the RF and XGBoost models were evaluated to ensure optimal configuration for each output. Also, variable influence was examined through permutation feature importance for ensemble models and statistical significance testing for linear regression, enabling interpretation and discussion of the influence of input variables on selected outputs.

A proper diet that provides balanced nutrients according to species, breed, age, sex, and purpose is essential for achieving optimal animal growth, development, and physiological function. In relation to this statement, the aim of this study was to quantitatively assess the effects of different dietary regimens on the morphometric characteristics of the small intestine of late puerperal rats, with a emphasis on the duodenum and jejunum. The research included measurements of the height and width of intestinal villi, and the depth of crypts, in order to determine specific morphological changes associated with the influence of the diet. The experiment involved 18 adult rats divided into three groups: the first group received standard commercial rat feed (control group), the second group was fed bakery products, and the third group was given a diet consisting exclusively of meat. Over a 49-day period, body weight, organ weights, and intestinal segment lengths were measured. Histological analyses of duodenum and jejunum samples were performed, alongside detailed morphometric assessment of the intestinal mucosa. Significant alterations in intestinal villi were observed. In the duodenum, the greatest villus height and width, as well as crypt depth, were observed in rats that had meat-based diet. Similarly, in the jejunum, rats from the same group exhibited the greatest villus height and crypt depth, while the mean measure of the villus width was almost identical to mean measure in rats that were fed with commercial food. These findings suggest that diet composition profoundly influences intestinal architecture and function. Overall, the study emphasizes the critical role of balanced nutrition in maintaining gastrointestinal health, systemic homeostasis, and optimal development, not only in laboratory animals but also in a broader veterinary and biomedical context.

INTRODUCTION Titanium fastener technology is increasingly adopted for use in minimally invasive valve surgery and open procedures to facilitate faster operative times and reproducible fastening of sutures. The objective of this study was to evaluate the technical feasibility and short-term safety of automated titanium fastener technology for securing sutures in heart valve repair and/or heart valve replacement procedures. METHODS The CRIMP study is a multicenter, prospective study that enrolled patients undergoing heart valve repair or replacement via open or minimally invasive approaches at three centers (n = 120). The primary endpoints were device success and prosthesis implantation time. RESULTS Mean age was 62.5 (SD 11.1) years, 33.3% of patients was female and the cohort was characterized as low surgical risk (EuroSCORE II 1.6%; SD 2.3%). Predominant valve disease was mitral regurgitation. Median procedure time was 178 (IQR 145-210) min. Median aortic cross-clamp time was 82 (IQR 59-107) min. The majority of procedures was minimally invasive (76.7%). In aortic valve procedures, median prosthesis implantation time (first stitch until last COR-KNOT placement) was 25 (IQR 18-46) min and in mitral valve procedures median implantation time was 50 (IQR 48-52) min. Device success was 100%, since no automated titanium fastener failed to grip or crimp appropriately. Median number of COR-KNOTS used was 14 (IQR 12-16) per valve. In-hospital mortality was 0.8%. All observed adverse events were considered unrelated to the device and there was no valve prosthesis dehiscence at 30-days of follow-up. CONCLUSION The use of automated titanium fastener technology for suture fixation in heart valve surgery was technically feasible and showed short-term safety in all patients, with no obvious device-related adverse events observed. Graphical abstract available for this article.  Clinical trial registration number: DRKS00038956.

This cross-sectional two-step study aimed to investigate the prevalence of eating disorder (ED) symptoms and associated mental health difficulties among elite athletes and an age-and sex matched comparison group. National team athletes were recruited through national sport federations, while participants in the comparison group were recruited via social media. A total of 408 participants (athletes n = 256; comparison group n = 152) (72.1% female), completed an anonymous online survey comprising validated psychometric measures of ED symptoms, exercise addiction, depression and psychological flexibility. Participants who screened positive for ED symptoms and providing contact details were invited to undergo a clinical assessment using the ED Examination Interview (EDE 17.0D). Overall, 25.0% (n = 102) reported ED symptoms, with a higher prevalence in the comparison group than among athletes (37.5% vs. 17.6%, p < 0.001). Interviews were completed by 78.0% (n = 18) of athletes and 47.0% (n = 16) of comparison group participants who screened positive for ED symptoms, confirming an ED diagnosis in all but one athlete. Furthermore, 18.6% (n = 76) reported clinically relevant symptoms of depression, and 22.8% (n = 93) of exercise addiction. Associations were observed between ED symptoms and symptoms of exercise addiction, depression, and psychological inflexibility in the total sample, among athletes and in the comparison group. ED symptoms were common among athletes and the comparison group and were associated with depressive symptoms, problematic exercise behaviour, and psychological inflexibility. To safeguard athlete health and optimise performance, national sports organisations should implement systematic preventive strategies, promote early identification, establish clear referral pathways, and ensure access to coordinated multidisciplinary care.

The electromagnetic coupling phase at the complex resonance pole is a fundamental property of nucleon excitations. However, its extraction from pion photoproduction data remains model-dependent, particularly for the Δ(1232) resonance, where modern multichannel analyses report helicity amplitude phases ranging from +3° to −18°. In this Letter, we present a largely model-independent geometric S-matrix formalism that provides a physical constraint for this ambiguity. By imposing Watson’s final-state interaction theorem, a direct consequence of S-matrix unitarity and time-reversal symmetry, on the real energy axis and performing an analytic continuation, we isolate the kinematic threshold barriers of the photoproduction (k·q) and elastic (q3) amplitudes. For the dominant M1+ multipole transition of the Δ(1232), our method yields a kinematically constrained prediction of ϕEM=−8.5°−1.0°+0.6° for the electromagnetic residue phase. This geometric constraint explains the numerical results of coupled-channel phenomenological fits, validating the extractions by the SAID and Bonn–Gatchina groups, and establishes a theoretical benchmark for evaluating resonance properties.