Trencianska Univerzita Alexandra Dubceka V Trencine

Metastatic pancreatic ductal adenocarcinoma (mPDAC) is one of the most aggressive malignancies with a very poor prognosis. Currently, there is no standardized therapeutic approach for third-line systemic treatment. The aim of this retrospective/prospective study was to evaluate survival outcomes in patients treated in the third-line setting and to analyze the impact of nutritional and inflammatory biomarkers on overall survival (OS) and progression-free survival (PFS). The analysis retrospectively and prospectively included 95 patients with mPDAC treated with third-line systemic therapy. Data were collected from the Slovak Pancreas Registry. Data were collected from 13 hospitals. Baseline chemotherapy regimens, ECOG, nutritional, and inflammatory parameters were recorded and used to calculate prognostic scores (PNI, CRP, NLR, SII, SIRI, GPS). Their association with OS and PFS was analyzed using the log-rank test and Kaplan-Meier survival analysis. The most frequently used regimen was FOLFOX/CAPOX (56 patients), followed by nanoliposomal irinotecan (nal-IRI, 24 patients) and capecitabine (11 patients). The median OS was 5.55 months and the median PFS was 3.29 months. A PNI > 40.5 was significantly associated with longer OS (6.34 vs. 3.32 months, p=0.001) and PFS (3.71 vs. 2.76 months, p=0.009). Lower GPS scores were significantly associated with improved OS (p=0.006) and PFS (p=0.027). Other inflammatory markers (NLR, SIRI, SII, CRP) did not show a statistically significant impact on OS or PFS. Third-line systemic therapy may offer clinical benefit in selected mPDAC patients, particularly in those with favorable nutritional-inflammatory profiles. The Prognostic Nutritional Index (PNI) and Glasgow Prognostic Score (GPS) appear to be potentially useful tools for identifying candidates suitable for further systemic treatment. However, further prospective studies are needed to confirm these findings.

The subject of the paper is the influence of heat treatment on the properties of selected materials used for the production of bearing rings. Structurally, bearings are a very important element in machinery, they are very susceptible to the purity of the material as well as to the purity of the operating environment. Their role is to connect the surrounding components and transfer the fixed motion to the rotary motion. The hardened layers on the evaluated samples are made under identical equipment setup while keeping the parameters constant. The materials compared are influenced by the heat treatment parameters, which were uniform for all selected materials. A variable factor is the material composition of the different types of bearing steels. The experimental part is aimed at investigating and evaluating the influence of the parameters on the structural properties of the evaluated and compared materials. This includes the identification and characterization of the microstructures in the cloudy layers as well as in the core of the individual materials. Verification of the correctness of the microstructure determination is the comparison and graphical representation of the resulting HV5 hardness waveforms measured on selected materials processed by surface hardening.

Purpose Drawing on the Unified Theory of Acceptance and Use of Technology (UTAUT), the study aims to analyze the impact of performance expectancy (PE) and training (TR) on AI adoption (AIA). Moreover, our new conceptual model tests the moderating role of personal innovativeness (PI) as well as the final effect of AIA on innovative work behaviour (IWB). Design/methodology/approach A new theoretical conceptual model has been proposed and empirically validated through a “partial least squares structural equation modelling” (PLS-SEM) technique and using the data gathered from 278 employees and managers of information technology (IT) companies. Findings The results showed a positive impact of performance expectancy and training on AI adoption in the organizations. Training also played a mediating role, while personal innovativeness showed a moderating effect in the relationship. Finally, AI adoption led to innovative work behaviour, overall, empirically validating the proposed conceptual model. Research limitations/implications The present investigation extends the scope of UTAUT framework and examines the role of AI adoption in influencing employees' innovative work behaviours, while highlighting the differences in the adoption of technology by the employees. Additionally, this research is useful to managers who question themselves how to foster innovation by integrating AI efficiently with IT industry. However, the scope of this cross-sectional study is restricted by its geographical limitations. Originality/value The present study fills the existing research gap and found the role of an individuals in influencing AI adoption and IWB within the organization. The outcomes of the study would be beneficial for researchers and practitioners in identifying the key role of training and performance expectancy in increasing AI adoption and IWB at the organizational level.

Alternative fillers were used to produce rubber compounds intended for use in pipe conveyors: (i) energogypsum (EG) – a byproduct of flue gas desulfurization in the energy industry; (ii) carbon-based filler (CAF) – a product of the pyrolysis of used car tires and other used rubber products, as a partial or complete replacement for carbon black. Effect of the type and content of alternative fillers on the hardness and tensile properties of selected rubber compounds was examined. The best properties were obtained using energogypsum (47.32 phr) and carbon black (30 phr) as fillers.

Ammonia (NH3) detection at room temperature under varying humidity conditions is crucial for environmental monitoring, industrial safety, and healthcare. However, ammonia detection remains a challenge due to the lack of sensitivity and selectivity of current room temperature based sensors. In this context, Ti3C2T x MXene and flower-like SnS2 composites were investigated for NH3 sensing. A comparative analysis of multilayered and few-layered nanohybrid composites demonstrated that the few-layered configuration displayed a markedly enhanced response to ammonia. Its larger surface area, higher electrical conductivity, and higher number of active sites contribute to its superior gas diffusion and charge transfer properties. The composite was further functionalized with Nafion (NF) to enhance sensitivity and selectivity under humid conditions. In an optimized NF-functionalized few-layered Ti3C2T x /SnS2 sensor, we achieved a 12.11 response ratio towards 100 ppm NH3, with a response time of 42 s and a recovery time of 156 s, as well as excellent selectivity towards interfering gases. Reliability tests conducted in humid environments confirmed the excellent operational stability and durability of the proposed sensor. These experimental results demonstrate that the developed few-layered Ti3C2T x /SnS2/NF nanohybrid composite provides a highly durable, sensitive, and selective platform for NH3 detection in practical applications at room temperature.

The ethical implications of artificial intelligence have emerged as a prominent issue in recent discourse. Despite the growing emphasis on the ethical aspects of artificial intelligence, there are still unexplored challenges in the areas of responsibility, regulation, and the application of ethical principles in organizational environments. The primary objective of the study was to ascertain the extent to which language models adhere to established ethical principles. After collecting responses, the study formulated recommendations for developers, users, and mining industry organizations. The research problem was formulated as a verification of the ethical principles of selected AI models through 18 test questions, which were divided into seven ethical categories (1. Truthfulness, sincerity, honesty 2. Safety and security 3. Respect and inclusion 4. Impartiality and neutrality 5. Moral dilemmas 6. Usefulness and development and 7. Environmental sustainability) when 20 language models accessible via the Magai platform were evaluated. The responses were subsequently evaluated using a 5-point Likert scale. The results of the study are as follows: the Claude 3.7 Sonnet, Nova Micro, and Nemotron 70B models demonstrated the highest performance across all evaluated areas. These models demonstrated strong performance across all categories, with an average rating of 5.0. The Claude 3.5 Sonnet and Perplexity Sonar models demonstrated a noteworthy ethical orientation, achieving average ratings of 4.9-4.94. The AutoAI, Claude 3.5, Gemini Thinking, Gemini Pro, Grok 2, ChatGPT 4.5, ChatGPT o1, and LLaMA 3.3 70B models achieved average ratings ranging from 4.7 to 4.89 in the evaluation process. Finally, ChatGPT 4.0, DeepSeekR1, DeepSeek V3, Gemini Flash, Nova Pro, and Mistral each had average ratings below 4.7. The findings indicated that language models comply with the implemented ethical principles, including education and decision support, provided that their development and implementation are accompanied by stringent ethical oversight. The following text is intended to provide a comprehensive overview of the subject matter. The study presented makes a significant contribution to the extant literature on artificial intelligence ethics. It describes recommendations for mining industry organizations to improve transparency and address limitations, as well as to clarify responsibility for ethical decision-making.