Search Results (19,464)

We investigate whether China’s Value-Added Tax (VAT) Credit Refund policy influences firms’ sustainable development capability (SDC), which reflects innovation-driven growth and green development. Exploiting the 2018 implementation of the VAT Credit Refund policy as a quasi-natural experiment, we employ a difference-in-differences (DID) approach and find causal evidence that the policy significantly enhances firms’ SDC. This suggests that fiscal instruments like VAT refunds are valued by firms as drivers of long-term sustainable and high-quality development. Our mediating analyses further reveal that the policy promotes firms’ SDC by strengthening artificial intelligence (AI) capabilities and facilitating intelligent transformation. This mechanism “AI Capability Building—Intelligent Transformation” aligns with the socio-technical systems theory (STST), highlighting the interactive evolution of technological and social subsystems in shaping firm capabilities. The heterogeneity analyses indicate that the positive effect of VAT Credit Refund policy on SDC is more pronounced among small-scale and non-high-tech firms, firms with lower perceived economic policy uncertainty, higher operational diversification, lower reputational capital, and those located in regions with a higher level of marketization. We also find that the policy has persistent long-term effects, with improved SDC associated with enhanced ESG performance and green innovation outcomes. Our findings have important implications for understanding the SDC through the lens of STST and offer policy insights for deepening VAT reform and promoting intelligent and green transformation in China’s enterprises. Full article

Background and Objectives: Periodontitis is a chronic inflammatory disease that leads to progressive destruction of periodontal tissues and remains a significant global health burden. While conventional therapies such as scaling and root planning offer short-term improvements, they often fall short in maintaining long-term microbial control, underscoring the need for adjunctive strategies. This study evaluated the clinical and microbiological effects of a novel essential oil (EO)-based gel—SmartGel OV—formulated with Origanum vulgare. Materials and Methods: Thirty adults with periodontitis were enrolled in a 4-month observational study, during which SmartGel OV was applied daily via gingival massage. Clinical outcomes and bacterial profiles were assessed through probing measurements and real-time PCR analysis. Additionally, a pilot AI-based tool was explored as a supplemental method to monitor inflammation progression through intraoral images. Results: Significant reductions were observed in Fusobacterium nucleatum and Capnocytophaga spp., accompanied by improvements in clinical markers, including probing depth, bleeding on probing, and plaque index. The AI framework successfully identified visual inflammation changes and supported early detection of non-responsiveness. Conclusions: SmartGel OV demonstrates promise as a natural adjunctive treatment for periodontitis and AI monitoring was included as an exploratory secondary tool to assess feasibility for future remote tracking. Full article

The rise in illicit financial activities across the South Africa–Zimbabwe corridor, with an estimated annual loss of $3.1 billion demands advanced AI solutions to augment traditional detection methods. This study introduces FALCON, a groundbreaking hybrid transformer–GNN model that integrates temporal transaction analysis (TimeGAN) and graph-based entity mapping (GraphSAGE) to detect illicit financial flows with unprecedented precision. By leveraging data from South Africa’s FIC, Zimbabwe’s RBZ, and SWIFT, FALCON achieved 98.7%, surpassing Random Forest (72.1%) and human auditors (64.5%), while reducing false positives to 1.2% (AUC-ROC: 0.992). Tested on 1.8 million transactions, including falsified CTRs, STRs, and Ethereum blockchain data, FALCON uncovered $450 million laundered by 23 shell companies with a cross-border detection precision of 94%, directly mitigating illicit financial flows in Southern Africa. For regulators, FALCON met FAFT standards, yielding 92% court admissibility, and its GDPR-compliant design (ε = 1.2 differential privacy) met stringent legal standards. Deployed on AWS Graviton3, FALCON processed 2 million transactions/second at $0.002 per 1000 transactions, demonstrating real-time scalability, making it cost-effective for financial institutions in emerging markets. As the first AI framework tailored for Southern Africa’s financial ecosystems, FALCON sets a new benchmark for ethical AML solutions in emerging economies with immediate applicability to CBDC supervision. The transparent validation of publicly available data underscores its potential to transform global financial crime detection. Full article

The integration of AI into online platforms is reshaping consumer experience and behavior. While existing research has largely focused on the role of AI in search services and experience services, few studies have examined the role of AI in the context of credence services. This study fills this gap by investigating an AI-powered preview tool in the context of online medical aesthetic platforms. Specifically, this study investigates how the AI Magic Mirror influences consumer purchase intentions in medical aesthetic services. Using secondary data analysis and two experimental studies, we examine the main effects, as well as mediation and moderation effects. The findings consistently demonstrate that the AI Magic Mirror significantly increases consumer purchase intentions. This relationship is positively mediated by perceived value and negatively mediated by perceived risk. In addition, the main effect is stronger for procedures with higher fit uncertainty and is more pronounced for those with lower popularity. These results provide theoretical insights into AI application in credence service contexts and offer practical implications for the design of AI-enhanced online service platforms. Full article

In resistance spot welding (RSW), the total electrical resistance (dynamic resistance) as the sum of bulk and contact resistance is a key variable. Both of these respective resistances influence the welding result, but the exact ratio to the total resistance of a real existing sheet is not known. Due to the high scatter in the RSW of aluminum alloys compared to steel, it is of interest to be able to explicitly determine the individual resistance components in order to gain a better understanding of the relationship between the initial state and the lower reproducibility of aluminum welding in the future. So far, only the total resistance and the bulk resistance could be determined experimentally. Due to the different sample shapes, it was not possible to consistently determine the contact resistance from the measurements. In order to realize this, a method was developed that contains the following innovations with the aid of simulation: determination of the absolute bulk resistance at room temperature (RT), determination of the absolute contact resistance at RT and determination of the ratio of bulk and contact resistance. This method makes it possible to compare the resistances of the bulk material and the surface in the initial state quantitatively. This now allows the comparison of batches regarding the surface resistance, especially for welding processes. For the aluminum sheets (EN AW-5182-O, EN AW-6014-T4) investigated, the method showed that the contact resistance dominates and the bulk resistance is less than 20%. These data can also be used to make predictions about the weldability of the alloy using artificial intelligence (AI). If experimental data are available, the method can also be applied to higher temperatures. Full article

This paper presents a scientometric review that provides a quantitative perspective on the evolution of Automation in Construction Journal (AICJ) research, emphasizing its developmental paths and emerging trends. The study aims to analyze the journal’s growth and citation impact over time. It also seeks to identify the most influential publications and the cooperation patterns among key contributors. Furthermore, the study explores the journal’s primary research themes and their evolution. Accordingly, 4084 articles were identified using the Web of Science (WoS) database and subjected to a multistep analysis using VOsviewer version 1.6.18 and Biblioshiny as software tools. First, the growth and citation of the publications over time are inspected and evaluated, in addition to ranking the most influential documents. Second, the co-authorship analysis method is applied to visualize the cooperation patterns between countries, organizations, and authors. Finally, the publications are analyzed using keyword co-occurrence and keyword thematic evolution analyses, revealing five major research clusters: (i) foundational optimization, (ii) deep learning and computer vision, (iii) building information modeling, (iv) 3D printing and robotics, and (v) machine learning. Additionally, the analysis reveals significant growth in publications (54.5%) and citations (78.0%) from 2018 to 2023, indicating the journal’s increasing global influence. This period also highlights the accelerated adoption of digitalization (e.g., BIM, computational design), increased integration of AI and machine learning for automation and predictive analytics, and rapid growth of robotics and 3D printing, driving sustainable and innovative construction practices. The paper’s findings can help readers and researchers gain a thorough understanding of the AICJ’s published work, aid research groups in planning and optimizing their research efforts, and inform editorial boards on the most promising areas in the existing body of knowledge for further investigation and development. Full article

One of the most popular and innovative methods to analyse signals is by using Ordinal Patterns (OPs). The OP encoding is based on transforming a (univariate) signal into a symbolic sequence of OPs, where each OP represents the number of permutations needed to order a small subset of the signal’s magnitudes. This implies that OPs are conceptually clear, methodologically simple to implement, and robust to noise, and that they can be applied to short signals. Moreover, they simplify the statistical analyses that can be carried out on a signal, such as entropy and complexity quantifications. However, because of the relative ordering, information about the magnitude of the signal at each timestamp is lost—this being one of the major drawbacks of this method. Here, we propose a way to use the signal magnitudes discarded in the OP encoding as a complementary variable to its permutation entropy. To illustrate our approach, we analyse synthetic trajectories from logistic and Hénon maps—with and without added noise—and real-world signals, including intracranial electroencephalographic recordings from rats in different sleep-wake states and frequency fluctuations in power grids. Our results show that, when complementing the permutation entropy with the variability in the signal magnitudes, the characterisation of these signals is improved and the results remain explainable. This implies that our approach can be useful for feature engineering and improving AI classifiers, as typical machine learning algorithms need complementary signal features as inputs to improve classification accuracy. Full article

This paper presents a procedural method aimed at protecting maritime critical infrastructure, which is essential for the functioning of developed nations. A novel approach, developed by the authors, is introduced—focusing on the behavioral analysis of vessels to enable early identification of suspicious maritime activity and to prevent damage or destruction to key infrastructure elements. An integrated system is proposed, combining real-time electronic surveillance with continuous access to and analysis of data from both national and international databases. Drawing inspiration from medical sciences, a screening-based methodology has been developed. Data on vessels collected from various sources are processed according to the criteria adopted by the authors, using a multi-criteria decision analysis (MCDA) approach. MCDA is a decision-support method that considers multiple criteria simultaneously. It allows for the comparison and evaluation of different options, even when they are difficult to compare directly. This characteristic is used to select high-risk vessels for further monitoring. An initial classification of a vessel as suspicious does not constitute proof of criminal activity but rather serves as a trigger for further coordinated actions. Data on vessels is collected from the AIS (automatic identification system) and platforms that store vessel history. The AIS is a powerful tool that processes parameters such as a ship’s speed and course. This article presents sample results from surveillance and pre-selection analyses using the AIS, followed by a multi-criteria assessment of the behavior of vessels identified through this process. The results are presented both graphically and numerically. The authors conducted several scenarios, analyzing different groups of vessels. Based on this analysis, recommendations were developed for the interpretation of the findings. Full article

Virtual AI-driven streamers have been gradually used in live commerce, and they may affect consumer impulsive purchase intention. Drawing on the stimulus–organism–response (S-O-R) model, this research examined consumer impulsive purchase intention in virtual AI streaming. Based on survey data from 411 predominantly young and educated virtual AI streaming users recruited through snowball sampling, we found that perceived responsiveness, perceived likeability, perceived expertise, and perceived anthropomorphism of virtual AI streamers are associated with trust and flow experience, both of which predict consumers’ impulsive purchase intentions. The fsQCA identified two paths that lead to impulsive purchase intention. The results imply that live streaming platforms need to engender consumers’ trust and flow experience in order to increase their impulsive purchase intention. Full article

The present study explores the application of RNNs for the prediction and propagation of flood waves along a section of the Bârsa River, Romania, as a fast alternative to classical hydraulic models, aiming to identify new ways to alert the population. Five neural architectures were analyzed as follows: S-RNN, LSTM, GRU, Bi-LSTM, and Bi-GRU. The input data for the neural networks were derived from 2D hydraulic simulations conducted using HEC-RAS software, which provided the necessary training data for the models. It should be mentioned that the input data for the hydraulic model are synthetic hydrographs, derived from the statistical processing of recorded floods. Performance evaluation was based on standard metrics such as NSE, $R^2$ MSE, and RMSE. The results indicate that all studied networks performed well, with NSE and $R^2$ values close to 1, thus validating their capacity to reproduce complex hydrological dynamics. Overall, all models yielded satisfactory results, making them useful tools particularly the GRU and Bi-GRU architectures, which showed the most balanced behavior, delivering low errors and high stability in predicting peak discharge, water level, and flood wave volume. The GRU and Bi-GRU networks yielded the best performance, with RMSE values below 1.45, MAE under 0.3, and volume errors typically under 3%. On the other hand, LSTM architecture exhibited the most significant instability and errors, especially in estimating the flood wave volume, often having errors exceeding 9% in some sections. The study concludes by identifying several limitations, including the heavy reliance on synthetic data and its local applicability, while also proposing solutions for future analyses, such as the integration of real-world data and the expansion of the methodology to diverse river basins thus providing greater significance to RNN models. The final conclusions highlight that RNNs are powerful tools in flood risk management, contributing to the development of fast and efficient early warning systems for extreme hydrological and meteorological events. Full article

Generative AI tools such as ChatGPT are reshaping educational practice, yet first-year teacher-education students often lack the prompt-engineering skills and confidence required to use them responsibly. This pilot study examined whether a concise three-session clinic on prompt engineering could simultaneously boost AI literacy and reduce technology anxiety in prospective teachers. Forty-five freshmen in a Peruvian teacher-education program completed validated Spanish versions of a 12-item AI-literacy scale and a 12-item technology-anxiety scale one week before and after the intervention; normality-checked pre–post differences were analysed with paired-samples t-tests, Cohen’s d, and Pearson correlations. AI literacy rose by 0.70 ± 0.46 points (t (44) = −6.10, p < 0.001, d = 0.91), while technology anxiety fell by 0.58 ± 0.52 points (t (44) = −3.82, p = 0.001, d = 0.56); individual gains were inversely correlated (r = −0.46, p = 0.002). These findings suggest that integrating micro-level prompt-engineering clinics in the first semester can help future teachers engage critically and comfortably with generative AI and guide curriculum designers in updating teacher-training programs. Full article

Hodgkin’s Lymphoma (HL) affects approximately 8500 individuals annually in the United States. The 5-year relative survival rate has improved to 88.5%, driven by transformative advances in immunotherapy and precision oncology. The integration of Brentuximab vedotin (BV) and immune checkpoint inhibitors (ICIs) has redefined treatment paradigms. The phase III SWOG S1826 trial established nivolumab plus doxorubicin, vinblastine, and dacarbazine (N + AVD) as an emerging new standard for advanced-stage HL, achieving a 2-year progression-free survival (PFS) of 92% compared to 83% for BV plus AVD (HR 0.48, 95% CI: 0.33–0.70), with superior safety, particularly in patients over 60. In relapsed/refractory HL, pembrolizumab outperforms BV, with a median PFS of 13.2 versus 8.3 months (HR 0.65, 95% CI: 0.48–0.88), as demonstrated in the KEYNOTE-204 trial. Emerging strategies, including novel ICI combinations, minimal residual disease (MRD) monitoring via circulating tumor DNA (ctDNA), and artificial intelligence (AI)-driven diagnostics, promise to further personalize therapy. This review synthesizes HL’s epidemiology, pathogenesis, diagnostic innovations, and therapeutic advances, highlighting the role of precision medicine in addressing unmet needs and disparities in HL care. Full article

Introduction: A validated clinical decision tool predictive of favorable functional outcomes following endovascular thrombectomy (EVT) in acute ischemic stroke (AIS) remains elusive. We performed a retrospective case series of patients at our regional Comprehensive Stroke Center, over a four-year period, who have undergone EVT to elucidate patient characteristics and factors associated with a favorable functional outcome after EVT. Methods: We reviewed all cases of EVT at our institution between February 2018 and February 2022 in the extended time window from 6–24 h. Demographic, clinical, imaging, and procedure co-variates were included. A favorable clinical outcome was defined as a modified Rankin scale of 0–2. We included patients with M1 or internal carotid artery occlusion treated with EVT within 6–24 h after symptom onset. We used a univariate and multivariate logistic regression analysis to identify patient factors associated with a favorable clinical outcome at 90 days. Results: Our study included evaluation of 121 patients who underwent EVT at our comprehensive stroke center. Our analysis demonstrates that a higher recanalization score based on the modified Thrombolysis In Cerebral Infarction (mTICI) scale (2B-3) was a strong indicator of a favorable outcome (OR 7.33; CI 2.06–26.07; p = 0.0021). Our data also showed that a higher baseline National Institutes of Health Stroke Scale (NIHSS) score (p = 0.0095) and the presence of pre-existing hypertension (p = 0.0035) may also be predictors of an unfavorable outcome (mRS > 2) per our multivariate analysis. Conclusion: Patients without pre-existing hypertension had more favorable outcomes following EVT in the expanded time window. This is consistent with other multicenter data in the expanded time window that demonstrates greater odds of a poor outcome with elevated pre-, peri-, and post-endovascular-treatment blood pressure. Our data also demonstrate that the mTICI score is a strong predictor of favorable outcome, even after controlling for other variables. A lower baseline NIHSS at the time of thrombectomy may also indicate a favorable outcome. Furthermore, the presence of clinical or radiographic mismatch based on the Alberta Stroke Program Early Computed Tomography Score (ASPECTS) and NIHSS per DAWN and DEFUSE-3 criteria did not emerge as a predictor of favorable outcome, which is congruent with recent randomized controlled trials and meta-analyses. Full article

The advancement of artificial intelligence (AI), deep learning, and radiomics has introduced novel methodologies for the detection, classification, prognosis, and treatment evaluation of pancreatic ductal adenocarcinoma (PDAC). As the integration of AI into medical imaging continues to evolve, its potential to enhance early detection, refine diagnostic precision, and optimize treatment strategies becomes increasingly evident. However, despite significant progress, various challenges remain, particularly in terms of clinical applicability, generalizability, interpretability, and integration into routine practice. Understanding the current state of research is crucial for identifying gaps in the literature and exploring opportunities for future advancements. This literature review aims to provide a comprehensive overview of the existing studies on AI applications in PDAC, with a focus on disease detection, classification, survival prediction, treatment response assessment, and radiogenomics. By analyzing the methodologies, findings, and limitations of these studies, we aim to highlight the strengths of AI-driven approaches while addressing critical gaps that hinder their clinical translation. Furthermore, this review aims to discuss future directions in the field, emphasizing the need for multi-institutional collaborations, explainable AI models, and the integration of multi-modal data to advance the role of AI in personalized medicine for PDAC. Full article

The transition to smart, zero-carbon cities relies on advanced, sustainable energy solutions, with artificial intelligence (AI) playing a crucial role in optimizing renewable energy management. This study evaluates state-of-the-art AI models for solar power forecasting, emphasizing accuracy, reliability, and environmental sustainability. Using operational data from Benban Solar Park in Egypt and Sakaka Solar Power Plant in Saudi Arabia, two of the world’s largest solar installations, the research highlights the effectiveness of hybrid AI techniques. The hybrid Convolutional Neural Network–Long Short-Term Memory (CNN-LSTM) model outperformed other models, achieving a Mean Absolute Percentage Error (MAPE) of 2.04%, Root Mean Square Error (RMSE) of 184, Mean Absolute Error (MAE) of 252, and R² of 0.99 for Benban, and an MAPE of 2.00%, RMSE of 190, MAE of 255, and R² of 0.98 for Sakaka. This model excels at capturing complex spatiotemporal patterns in solar data while maintaining low computational CO₂ emissions, supporting sustainable AI practices. The findings demonstrate the potential of hybrid AI models to enhance the accuracy and sustainability of solar power forecasting, thereby contributing to efficient, resilient, and zero-carbon urban environments. This research provides valuable insights for policymakers and stakeholders aiming to advance smart energy infrastructure. Full article