The cement industry is under constant pressure to reduce its environmental footprint while ensuring economic competitiveness and technological reliability. One of the most effective strategies to achieve this goal is the substitution of traditional raw materials with alternative ones derived from industrial (by)products, waste, or secondary resources. This paper presents a structured methodology for the selection and evaluation of potential raw materials for clinker production. The proposed approach integrates four key criteria: physical compatibility, which determines whether the raw material can be handled by existing processing equipment; chemical compatibility, which ensures compliance with clinker quality requirements; environmental compliance, which assesses adherence to local and international environmental regulations; and economic viability, including the costs of material acquisition, processing, equipment adaptation, and CO2 emissions associated with the raw mix. The research procedure involves initial communication with suppliers, visual inspection of the material, laboratory analysis (chemical and environmental), raw mix modelling, and full economic evaluation. If at any stage the material fails to meet the required criteria, feedback is provided to the supplier, avoiding unnecessary costs and efforts. Results indicate that this integrated methodology offers a systematic and transparent making of decision framework that can accelerate the acceptance of alternative raw materials, improve resource efficiency, and contribute to sustainable cement production.

Background: The study investigates newly developed composite materials with advanced filler technology and modified resin matrices, designed to enhance esthetic quality, clinical efficiency, and mechanical properties. This study evaluated the effect of two light-curing protocols—a conventional low-voltage (LV) protocol (10 s at 1200 mW/cm2) and a high-voltage (HV) protocol (3 s at 3000 mW/cm2)—on the microhardness (MH), bottom/top MH ratio, and the correlation between filler content (wt%, vol%) and MH of bulk-fill resin-based composites (RBCs). Four RBCs were tested: Tetric PlusFill (TPF), Tetric Plus Flow(TPFW), Tetric PowerFill (PFL), and Tetric PowerFlow (PFW). Materials and Methods: Samples were fabricated in the laboratory using specially designed cylindrical molds (diameter = 8 mm, height = 4 mm). Initial MH was measured on the top and bottom surfaces of composite specimens 24 h after light curing using a digital microhardness tester (QNESS 60 M EVO, ATM Qness GmbH, Mammelzen, Germany). The correlation between the filler content (wt%, vol%) and the MH of the RBCs was tested. For the calculation of depth-dependent curing effectiveness, the bottom/top ratio for initial MH was used. Conclusions: The MH of bulk-fill RBCs was found to be influenced by both material composition and the applied light-curing protocol. An increase in filler content resulted in higher MH values under both protocols, with the filler volume fraction exhibiting a stronger correlation than the weight fraction. While both flowable and sculptable Tetric Plus composites exhibited higher MH values under the HV protocol, Tetric Power composites demonstrated greater initial hardness under LV protocol. The flowable composite PFW showed the most pronounced reduction in MH under HV curing. The bottom/top MH ratio exceeded 80% in all tested materials, confirming adequate polymerization throughout the composite layers.

Segmentation of brain tissue is an essential task in medical image analysis, particularly in neuroimaging and disease diagnosis. This study evaluates and compares three major segmentation approaches in the ISBR18 dataset: atlas-based methods, machine learning techniques, and deep learning architectures. The atlas-based Majority Voting method achieved the highest performance within its category with a dice similarity coefficient of 0.8477, utilizing anatomical templates for segmentation. Among machine learning techniques, K-means clustering demonstrated robust performance with 96% classification accuracy, offering computational efficiency despite limitations in spatial resolution. The deep learning U-Net model trained for binary segmentation achieved 93% accuracy, benefiting from its encoder-decoder architecture for precise boundary detection. While traditional atlas-based approaches provide robust anatomical consistency and machine learning methods offer computational advantages, deep learning models show promise in handling complex segmentation tasks. Future research could integrate these approaches to enhance segmentation performance in the ISBR18 dataset and lead to more accurate and reliable brain tissue segmentation for clinical applications.

Background Fermented foods are an integral part of diets worldwide, and emerging epidemiological data suggest their significant beneficial health effects. However, assessing their intake is challenging since many of these foods are sporadically and/or locally consumed, hence current traditional nutritional assessment tools lack the specificity to capture this variability. To address this gap, the Fermented Food Frequency Questionnaire (3FQ) was developed and this study aimed to evaluate its relative validity and repeatability across European regions. Methods In the validation study of the 3FQ, 12,646 adult participants were recruited across four European regions to assess consumption of sixteen major fermented food groups. Repeatability was assessed by administering the 3FQ twice, ~6 weeks apart, to a subset of participants (n = 2,315). Validity was evaluated using 24-h dietary recalls (24 h). Statistical analyses included Spearman's rank correlation coefficients and Intra-Class Correlation coefficients (ICC) for repeatability, and Bland-Altman plots for validity. Results Results showed high repeatability, overall and by region, for estimated quantities and frequencies of consumption for most of the fermented food groups (from 0.4 to 1.0), with a few exceptions for infrequently consumed items (e.g., fermented fish). Validity assessment via Bland-Altman plots revealed excellent agreement between the 3FQ and 24 h for most of the food groups, with over 90% of values falling within the agreement interval. Notably, fermented dairy products, coffee, and bread categories showed the strongest agreement (>95%). Conclusion The 3FQ is a robust and reliable tool for estimating the consumption of diverse fermented food groups across four European regions and provides valid estimates of the frequency and quantity of intake for specific fermented foods. The 3FQ could be a valuable instrument for epidemiological research aiming to elucidate associations between certain fermented foods and health parameters in European diets.

Contemporary approaches to teaching children’s literature and music education are increasingly replacing traditional pedagogical methods. The emergence of artificial intelligence (AI) has made these processes more dynamic and complex, presenting both opportunities and challenges for educators and pupils. This development raises important questions about teachers’ readiness to adopt innovative methodologies and pupils’ receptiveness to deeper learning and improved outcomes through AI-enhanced instruction. Children’s literature and music possess the capacity to educate, nurture, and heal. Their interdisciplinary nature provides primary school teachers with a rich foundation for creative and integrative teaching strategies, particularly relevant in the age of AI. This paper advocates for the deliberate integration of AI and healing education into the teaching of children’s literature and music at the primary level. It proposes that such integration can be achieved through creative and interdisciplinary applications of AI tools, alongside bibliotherapeutic and musicotherapeutic methods. The central methodological framework employed is mood mapping. The study examines the use of the Donna AI Song Generator within healing education, aiming to identify optimal strategies for both teachers and pupils. These findings may inform the development of diverse teaching methodologies and offer insights into the creative use of AI in interdisciplinary primary education. Additionally, the paper introduces an innovative conceptual framework – the Bibliotherapy and Musicotherapy Questionnaire (BMQ) – proposed as a theoretical model for future implementation in primary education settings. Although conceptual and theoretical in nature, the study is grounded in extensive practical teaching experience and the integration of AI tools, particularly within healing education. The BMQ model demonstrates adaptability to diverse instructional contexts and age groups, offering potential for future empirical validation and practical classroom application. Ultimately, this research highlights the transformative potential of AI in fostering holistic, creative, and therapeutic learning environments in primary education.

Background In the dot perspective taking task – a task commonly used to assess implicit mentalizing - participants are typically slower in judging how many dots they see when there is a difference in the number of dots seen by themselves and a centrally placed avatar than when both perspectives align. This finding has been termed the ‘altercentric interference’ effect and taken as evidence that participants spontaneously and automatically compute the avatar’s perspective. In this study, we focus on one line of critiques regarding the interpretation that the altercentric interference effect is automatic, namely by assessing whether the effect is purely stimulus-driven. Specifically, we tested the proposal that for the altercentric interference effect to emerge, participants must first focus their attention to the avatar, which then directs their attention to the dots and that this is achieved by a social word prompt (typically ‘YOU’) inducing a social mindset that then draws the participants’ attention to the avatar once the avatar and the dots appear. Methods We tested two groups of participants: one with a ‘YOU’ prompt and one with a non-social ‘NOW’ prompt. The semantics of both prompts were irrelevant because information about the colour of the dots participants needed to judge was presented through the ink colour of the prompt, not its text. Results Our results revealed no statistically significant difference in the altercentric interference effect between groups and our exploratory analyses showed that this was due to the altercentric interference effect being present in both groups. Conclusions Our findings do not provide empirical support for the hypothesis that the word prompt used in typical dot perspective tasking tasks promotes a social mindset that leads to the altercentric interference effect, however they may be aligned with the hypothesis that the effect requires participants’ attention to be drawn to the avatar.

Autonomic nervous system (ANS) dysfunction has emerged as a central feature of post-infectious syndromes, including post-COVID syndrome (PCS), chronic fatigue syndrome (CFS), and late-stage Lyme disease. This cross-sectional study included 1036 patients evaluated in the Neurocardiological Laboratory of the Institute for Cardiovascular Diseases “Dedinje,” divided into four groups: PCS, CFS after COVID-19, CFS of insidious onset, and Lyme disease. All patients underwent head-up tilt testing (HUTT), and serological testing was performed in accredited institutions. The Lyme disease group demonstrated the highest prevalence of positive HUTT responses and a significantly greater frequency of orthostatic hypotension and syncope. Approximately 50–65% of patients in the PCS and Lyme groups were positive for IgM antibodies against at least one microorganism, with more than 10% showing positivity for three or more pathogens. Logistic regression analysis revealed that, beyond classical hemodynamic parameters, antibody status served as a significant predictor of HUTT outcomes, with specific associations identified for HSV-1, HHV-6, Coxiella burnetii, Toxoplasma gondii, and Borrelia spp. Multinomial regression further indicated that negative IgG antibodies, particularly to HSV-1 and VZV, predicted Lyme disease group membership. These findings support the hypothesis that ANS dysfunction in post-infectious syndromes may be driven by persistent or prior infections, highlighting the need for integrative diagnostic approaches.

Background: Testicular torsion scoring systems, based on a combination of clinical and imaging factors, have been developed to improve the diagnostic accuracy of testicular torsion in patients presenting with acute scrotum. This study aimed to validate and compare two current testicular torsion scores the Boettcher Alert Score (BAL) and the Testicular Workup for Ischemia and Suspected Torsion (TWIST)-in a retrospective cohort of pediatric patients with acute scrotum. Methods: We conducted a retrospective study of all pediatric patients admitted to our institution for acute scrotum between January 2010 and December 2022. Patients were categorized into the testicular torsion (TT) group and the non-testicular torsion (NTT) group. Collected data were used to calculate the scoring systems and perform statistical analyses. Results: A total of 241 patients were included, of whom 80 (33.2%) had testicular torsion. The mean age in the TT group was 13 years. The optimal individual cut-off value for the BAL score was >1 (sensitivity 90%, specificity 80.75%), and for the TWIST score >4 (sensitivity 82.5%, specificity 80.75%). A high-risk TWIST score >5 had a specificity of 80.75% and a negative predictive value (NPV) of 90.28%, while a BAL score of 4 showed a specificity of 98.48% and NPV of 94.2%. The area under the ROC curve was slightly higher for the BAL score (0.917; 95% CI, 0.875–0.949) than for the TWIST score (0.897; 95% CI, 0.851–0.932). The difference between the two scores was not statistically significant. Conclusion: The TWIST and BAL clinical scores have significant diagnostic value and may assist in the evaluation of testicular torsion in children. Both scores could be incorporated into a standardized approach for assessing pediatric acute scrotum, potentially reducing time to definitive diagnosis, and minimizing ischemia duration.

We investigate the neutrino sector in the framework of flavor deconstruction with an inverse-seesaw realization. This setup naturally links the hierarchical charged-fermion masses to the anarchic pattern of light-neutrino mixing. We determine the viable parameter space consistent with oscillation data and study the phenomenology of heavy neutral leptons (HNL) and lepton-flavor-violating (LFV) processes. Current bounds from direct HNL searches and LFV decays constrain the right-handed neutrino scale to a few TeV, while future $\mu \to e$ experiments will probe most of the region with $\Lambda \lesssim 10~\text{TeV}$. Among possible realizations, models deconstructing $\mathrm{SU}(2)_\mathrm{L} \times \mathrm{U}(1)_\mathrm{B-L}$ or $\mathrm{SU}(2)_\mathrm{L} \times \mathrm{U}(1)_\mathrm{R} \times \mathrm{U}(1)_\mathrm{B-L}$ are those allowing the lowest deconstruction scale.

Zeolites are particularly suitable adsorbents due to their pronounced ion-exchange capacity, high efficiency, stability, and the ability to be regenerated and reused multiple times. Their characteristic crystalline structure enables the exchange of sodium, potassium, calcium, and magnesium ions with heavy metal cations present in solution. For the successful application of zeolites under industrial conditions, a detailed understanding of the adsorption mechanisms and kinetics is essential, as it allows for process optimization and identification of key limiting factors. Experimental approaches typically involve varying the adsorbent mass and the initial concentration of the adsorbate in order to determine the optimal conditions for achieving maximum adsorption efficiency. A moisture content of 3.95% and ash content of 91.28% indicate high thermal and structural stability of the zeolite, while the presence of Na⁺ ions (0.2435 mmol g⁻¹) in the material confirms that cation exchange is the dominant mechanism. Adsorption of heavy metals was investigated in a batch reactor at initial concentrations of 10, 50, and 100 mg/L, at a constant temperature of 298 K, with stirring at 200 rpm for 60 minutes. The amount of adsorbed ions was found to increase with rising equilibrium concentrations in the solution. Metal ion concentrations were determined using atomic absorption spectrophotometry. The highest adsorption was observed for Cu(II) ions within 5 minutes, while Cr(III) and Ni(II) ions reached their maximum adsorption within 20 minutes. The experimental data fit best to the Langmuir isotherm model, and the adsorption efficiency followed the order: Cu(II) > Cr(III) > Ni(II).

This study examines the impact of an unintended fire at the Drava International plastic processing facility near Osijek, Croatia, on soil quality and the potential human health risks associated with agricultural soils within a 10 and 20 km radius. Surface soil samples (0–5 cm) were collected from ten locations within 10 km three days after the incident, and eight composite samples were taken from sites 10–20 km away 17 days’ post-event. Additionally, 18 control samples previously collected for soil fertility or quality monitoring were included for comparative analysis. In total, 36 agricultural soil samples were analyzed for pH, organic matter, total phosphorus, potassium, calcium, magnesium, and trace elements (Cr, Co, Ni, Cu, Zn, As, Pb). Eighteen post-fire samples were also analyzed for polycyclic aromatic hydrocarbons (PAHs), dioxins, and perfluoroalkyl substances (PFAS). Ecological risk was assessed using the pollution load index (PLI) and enrichment factor (EF), while human health risk was evaluated through the estimation of incremental lifetime cancer risk (ILCR) and individual carcinogenic risks (CRi) for As, Cr, Ni, and Pb. Results showed that concentrations of dioxins (TEQ LB and UB), dioxin-like PCBs, and non-dioxin-like PCBs in samples within 10 km were either below detection limits or present in trace amounts (4.0 × 10−6 mg/kg). PFAS compounds were not detected (<0.0005 mg/kg). The total concentration of non-dioxin-like PCBs ranged from 0.0023 to 0.0047 mg/kg, well below the maximum permissible levels. Post-fire contamination profiles revealed consistently higher PAH concentrations in the 0–10 km zone (mean 0.100 mg/kg) compared to the 10–20 km zone (mean 0.062 mg/kg). Twenty PLI values exceeded the threshold of 1 (range: 1.00–1.26), indicating moderate pollution, while the remaining values (PLI 0.82–0.99) suggested no pollution. EF values indicated minimal to moderate enrichment (EF < 2), supporting the conclusion that metal presence was predominantly geological with limited anthropogenic influence. All ILCR values for adults and children remained below the acceptable threshold of 1 × 10−4, indicating low carcinogenic risk under both pre- and post-fire conditions. CRi values followed a consistent decreasing trend across exposure pathways: ingestion > dermal absorption > inhalation.

Background and Objectives: Resilience and perceived social support are crucial factors influencing psychological well-being among breast cancer survivors. Understanding their levels and interrelations can inform psychosocial interventions aimed at improving survivorship outcomes. This study aimed to examine the relationship between resilience and perceived social support, to evaluate the psychometric properties of the applied scales, and to explore their associations with key sociodemographic factors among breast cancer survivors. Materials and Methods: A total of 193 women in clinical remission, at least six months post-primary treatment, were recruited from the General Hospital Sombor. Participants completed sociodemographic and clinical questionnaires, the Connor–Davidson Resilience Scale (CD-RISC-25), and the Multidimensional Scale of Perceived Social Support (MSPSS). Descriptive statistics, Pearson’s correlations, and group comparisons (t-tests and ANOVA) were conducted to assess the relationships among study variables and sociodemographic factors. Results: Participants demonstrated moderate resilience (57 ± 18), with Coping and Hardiness as the strongest domains and Optimism the lowest. Perceived social support was also moderate (4.65–4.82) across all domains, highest for family and significant others. Resilience and perceived social support were positively correlated (r = 0.616, p < 0.001), with Hardiness most strongly associated with overall resilience (r = 0.899). Support from a significant other was particularly linked to adaptability (r = 0.617). Participants living in urban areas and those with higher income reported significantly higher resilience and social support, though with low effect sizes. No other sociodemographic associations were observed. Conclusions: Breast cancer survivors in this Serbian cohort reported moderate resilience and social support, with a strong interrelationship between the two. These findings underscore the importance of strengthening social support networks as a potential pathway to enhance resilience and psychological well-being in cancer survivorship care.

Background: Artificial intelligence (AI), the overarching field that includes machine learning (ML) and its subfield deep learning (DL), is rapidly transforming clinical research by enabling the analysis of high-dimensional data and automating the output of diagnostic and prognostic tests. As clinical trials become increasingly complex and costly, ML-based approaches (especially DL for image and signal data) offer promising solutions, although they require new approaches in clinical education. Objective: Explore current and emerging AI applications in oncology and cardiology, highlight real-world use cases, and discuss the challenges and future directions for responsible AI adoption. Methods: This narrative review summarizes various aspects of AI technology in clinical research, exploring its promise, use cases, and its limitations. The review was based on a literature search in PubMed covering publications from 2019 to 2025. Search terms included “artificial intelligence”, “machine learning”, “deep learning”, “oncology”, “cardiology”, “digital twin”. and “AI-ECG”. Preference was given to studies presenting validated or clinically applicable AI tools, while non-English articles, conference abstracts, and gray literature were excluded. Results: AI demonstrates significant potential in improving diagnostic accuracy, facilitating biomarker discovery, and detecting disease at an early stage. In clinical trials, AI improves patient stratification, site selection, and virtual simulations via digital twins. However, there are still challenges in harmonizing data, validating models, cross-disciplinary training, ensuring fairness, explainability, as well as the robustness of gold standards to which AI models are built. Conclusions: The integration of AI in clinical research can enhance efficiency, reduce costs, and facilitate clinical research as well as lead the way towards personalized medicine. Realizing this potential requires robust validation frameworks, transparent model interpretability, and collaborative efforts among clinicians, data scientists, and regulators. Interoperable data systems and cross-disciplinary education will be critical to enabling the integration of scalable, ethical, and trustworthy AI into healthcare.