Search Results (6,050)

Three-dimensional medical images, such as those obtained from MRI scans, offer a comprehensive view that aids in understanding complex shapes and abnormalities better than 2D images, such as X-ray, mammogram, ultrasound, and 2D CT slices. However, MRI machines are often inaccessible in certain regions due to their high cost, space and infrastructure requirements, a lack of skilled technicians, and safety concerns regarding metal implants. A viable alternative is generating 3D images from 2D scans, which can enhance medical analysis and diagnosis and also offer earlier detection of tumors and other abnormalities. This systematic review is focused on Generative Adversarial Networks (GANs) for 3D medical image analysis over the last three years, due to their dominant role in 3D medical imaging, offering unparalleled flexibility and adaptability for volumetric medical data, as compared to other generative models. GANs offer a promising solution by generating high-quality synthetic medical images, even with limited data, improving disease detection and classification. The existing surveys do not offer an up-to-date overview of the use of GANs in 3D medical imaging. This systematic review focuses on advancements in GAN technology for 3D medical imaging, analyzing studies, particularly from the recent years 2022–2025, and exploring applications, datasets, methods, algorithms, challenges, and outcomes. It affords particular focus to the modern GAN architectures, datasets, and codes that can be used for 3D medical imaging tasks, so readers looking to use GANs in their research could use this review to help them design their study. Based on PRISMA standards, five scientific databases were searched, including IEEE, Scopus, PubMed, Google Scholar, and Science Direct. A total of 1530 papers were retrieved on the basis of the inclusion criteria. The exclusion criteria were then applied, and after screening the title, abstract, and full-text volume, a total of 56 papers were extracted from these, which were then carefully studied. An overview of the various datasets that are used in 3D medical imaging is also presented. This paper concludes with a discussion of possible future work in this area. Full article

Extreme sea conditions caused by tropical cyclones pose significant risks to coastal safety, infrastructure, and ecosystems. Although existing models have advanced in predicting Significant Wave Height (SWH), their performance in predicting Maximum Wave Height (MWH) remains limited, particularly in capturing rapid wave fluctuations and localized meteorological dynamics. This study proposes a novel MWH prediction framework that integrates high-resolution buoy observations with deep learning. A moored buoy deployed in the Qiongzhou Strait provides precise nearshore observations, compensating for limitations in reanalysis datasets. Light Gradient Boosting Machine (LightGBM) is employed for key feature selection, and a hybrid Bidirectional Temporal Convolutional Network-Bidirectional Gated Recurrent Unit (BiTCN-BiGRU) model is constructed to capture both short- and long-term temporal dependencies. The results show that BiTCN-BiGRU outperforms BiGRU, reducing MAE by 6.11%, 5.41%, and 14.09% for 1-h, 3-h, and 6-h forecasts. This study also introduces the Time Distortion Index (TDI) into MWH prediction as a novel metric for evaluating temporal alignment. This study offers valuable insights for disaster warning, coastal protection, and risk mitigation under extreme marine conditions. Full article

Global warming and the energy crisis are two significant challenges in the world. The prime sources of greenhouse gas emissions are the transportation and power generation sectors because they rely on fossil fuels. To overcome these problems, the world needs to adopt electric vehicles (EVs) and renewable energy sources (RESs) as sustainable solutions. The rapid evolution of electric mobility is largely driven by the development of EV charging infrastructures (EVCIs), which provide the essential support for large-scale EV adoption. As the number of CIs grows, the utility grid faces more challenges, such as power quality issues, power demand, voltage instability, etc. These issues affect the grid performance, along with the battery lifecycle of the EVs and the charging system. A charging infrastructure integrated with the RES-based microgrid (MG) is an effective way to moderate the problem. Also, these methods are about reframing how smart sustainable cities generate, distribute, and consume energy. MG-based CI operates on-grid and off-grid based on the charging demand and trades electricity with the utility grid when required. This paper presents state-of-the-art transportation electrification, MG classification, and various energy sources in the DC MG. The grid-integrated DC MG, international standards for EV integration with the grid, impacts of CI on the electrical network, and potential methods to curtail the negative impact of EVs on the utility grid are explored comprehensively. The negative impact of EV load on the voltage profile and power loss of the IEEE 33 bus system is analysed in three diverse cases. This paper also provides directions for further research on grid-integrated DC MG-based charging infrastructure. Full article

Australia has set a new climate target of reducing emissions by 62–70% below 2005 levels by 2035, with sustainable aviation fuel (SAF) central to achieving this goal. This review critically examines techno-economic analysis (TEA) and life cycle assessment (LCA) of Power-to-Liquid (PtL) electrofuels (e-fuels), which synthesize atmospheric CO₂ and renewable hydrogen (H₂) via Fischer-Tropsch (FT) synthesis. Present PtL pathways require ~0.8 kg of H₂ and 3.1 kg of CO₂ per kg SAF, with ~75% kerosene yield. While third-generation feedstocks could cut greenhouse gas emissions by up to 93% (as low as 8 gCO₂e/MJ), real world reductions have been limited (~1.5%) due to variability in technology rollout and feedstock variability. Integrated TEA–LCA studies demonstrate up to 20% energy efficiency improvements and 40% cost reductions, but economic viability demands costs below $3/kg. In Australia, abundant solar resources, vast transport networks, and supportive policy frameworks present both opportunities and challenges. This review provides the first comprehensive assessment of PtL-FT SAF for Australian conditions, highlighting that large-scale development will require technological advancement, feedstock development, infrastructure investment, and coordinated policy support. Full article

Launched in 2013, China’s Belt and Road Initiative (BRI) was originally devised to link East Asia and Europe through a network of physical and digital infrastructure. This article analyses the BRI’s development in the European context by offering a comparative analysis of 727 BRI and BRI-like projects within 46 European countries from 2005 to 2021. The analysis considers projects’ location, typology, status, and the main enterprises involved in each project. According to our results, there is a “two-speed Europe”. Indeed, while the vast majority of projects are included in the Digital Silk Road (e.g., telecommunication, transfer technology, data centre, 5G, fintech) and are located in North-Western Europe, traditional investments in infrastructure (e.g., ports, roads, railways, SEZ) are concentrated in South-Eastern Europe and the Balkan countries. While North-Western Europe is particularly concerned about cyber security and data protection issues, various South-Eastern European countries look favorably upon the development opportunities offered by the BRI. The BRI is clearly different from the Western approach to development (based on competition and economic liberalism) and integration (based on treaties). The BRI approach—including its platform, leveraging political flexibility, economic pragmatism, ability to mobilize resources, and ability to create synergies between state and business—could take advantage of the flaws of the European integration process. The BRI, with its strengths as well as weaknesses, represents an opportunity for the EU to understand the need for greater economic and political foresight, social cohesion, and economic flexibility to meet the development needs of its member countries. China, too, can draw inspiration from cooperating with EU countries on how to improve the reception of its investment initiatives by focusing on reciprocity, security guarantees, and protection of rights and the environment. Full article

The accelerated development of digital technologies during the 21st century has intensified requirements for Human Settlements (HS) infrastructure advancement in China’s maritime urban centers, driven by national objectives to forge a cohesive, technologically integrated state framework. This transformation has changed people’s work, learning, and entertainment patterns, leading to the rise in complex networks of pseudo human settlements (PHS). Traditional approaches to environmental research are insufficient for understanding the interactions between PHS and reality human settlements (RHS), which are interdependent and shape urban development. This study utilizes advanced methods such as the entropy weight method to determine indicator weights, the coupling coordination degree model to quantify the interaction intensity, the geo-detector to identify driving factors, and ArcGIS for spatial analysis to assess the interaction between PHS and RHS in 53 coastal cities from 2011 to 2022. The results show: (1) The coupling coordination degree rose initially but later declined, reflecting temporal differentiation; (2) The coordination of settlements varies across regions; (3) A migration trend from the northeast to southwest, with faster coordination improvement in the southwest; (4) Socio-economic development drives the coupling coordination, with big data technology enhancing the relationship. The findings guide sustainable urban development in coastal cities. Full article

This study examines cost and schedule deviations in secondary road infrastructure projects in Colombia, with a focus on the influence of public procurement characteristics. Despite the construction sector’s importance to national development, limited research has explored how procurement-related variables affect project performance. To address this gap, 149 completed road projects were analyzed using data from Colombia’s open procurement database, which provides publicly accessible, standardized information on contracting processes. A four-stage methodology was applied: data collection, exploratory analysis, bivariate analysis (including correlation and Kruskal–Wallis tests), and multivariate analysis using Random Forest and Bayesian networks. Schedule and cost deviations were used as dependent variables, with 17 independent variables. Results show that 81.9% of projects experienced some form of deviation, with a positive correlation between schedule and cost overruns. Significant factors were identified across different stages of the project life cycle. Variables significant for both deviations include the number of bidders, the number of valid bidders, the estimated cost, the final cost, the project intensity, and the type of award process. The findings provide data-driven arguments to improve award processes and support more informed planning of future projects, helping public entities reduce deviations and enhance the outcome of their infrastructure. Full article

The Blockchain-Based Medical Insurance Transaction System (BMIT) developed in this study addresses key issues in traditional medical insurance—information silos, data tampering, and privacy breaches—through innovative blockchain architectural design and technical infrastructure reconstruction. Built on a consortium blockchain architecture with FISCO BCOS (Financial Blockchain Shenzhen Consortium Blockchain Open Source Platform) as the underlying platform, the system leverages FISCO BCOS’s distributed ledger, granular access control, and efficient consensus algorithms to enable multi-stakeholder on-chain collaboration. Four node roles and data protocols are defined: hospitals (on-chain data providers) generate 3D coordinate hashes of medical data via an algorithmically enhanced Bloom Filter for on-chain certification; patients control data access via blockchain private keys and unique parameters; insurance companies verify eligibility/claims using on-chain Bloom filters; the blockchain network stores encrypted key data (public keys, Bloom filter coordinates, and timestamps) to ensure immutability and traceability. A 3D-enhanced Bloom filter—tailored for on-chain use with user-specific hash functions and key control—stores only 3D coordinates (not raw data), cutting storage costs for 100 records to 1.27 KB and reducing the error rate to near zero ($1.77\%$ lower than traditional schemes for 10,000 entries). Three core smart contracts (identity registration, medical information certification, and automated verification) enable the automation of on-chain processes. Performance tests conducted on a 4-node consortium chain indicate a transaction throughput of 736 TPS (Transactions Per Second) and a per-operation latency of 181.7 ms, which meets the requirements of large-scale commercial applications. BMIT’s three-layer design (“underlying blockchain + enhanced Bloom filter + smart contracts”) delivers a balanced, efficient blockchain medical insurance prototype, offering a reusable technical framework for industry digital transformation. Full article

Water distribution networks (WDNs) are characterized by complex challenges in management and optimization, especially in ensuring efficiency, reducing losses, and maintaining infrastructure performances. The recent advancements in Artificial Intelligence (AI) techniques based on Large Language Models, particularly ChatGPT 4.0 (a chatbot based on a generative pre-trained model), offer potential solutions to streamline these processes. This study investigates the ability of ChatGPT to perform the clustering phase of WDN partitioning, a critical step for dividing large networks into manageable clusters. Using a real Italian network as a case study, ChatGPT was prompted to apply several clustering algorithms, including k-means, spectral, and hierarchical clustering. The results show that ChatGPT uniquely adds value by automating the entire workflow of WDN clustering—from reading input files and running algorithms to calculating performance indices and generating reports. This makes advanced water network partitioning accessible to users without programming or hydraulic modeling expertise. The study highlights ChatGPT’s role as a complementary tool: it accelerates repetitive tasks, supports decision-making with interpretable outputs, and lowers the entry barrier for utilities and practitioners. These findings demonstrate the practical potential of integrating large language models into water management, where they can democratize specialized methodologies and facilitate wider adoption of WDN managing strategies. Full article

Accurate prediction of structural displacements in hydropower stations is essential for the safety and long-term stability of large-scale water-related infrastructure. To address this challenge, this study proposes an AI-assisted monitoring framework that integrates Convolutional Neural Networks (CNNs) for spatial feature extraction with Gated Recurrent Units (GRUs) for temporal sequence modeling. The framework leverages long-sequence prototype monitoring data, including reservoir level, temperature, and displacement, to capture complex spatiotemporal interactions between environmental conditions and dam behavior. A parameter optimization strategy is further incorporated to refine the model’s architecture and hyperparameters. Experimental evaluations on real-world hydropower station datasets demonstrate that the proposed CNN–GRU model outperforms conventional statistical and machine learning methods, achieving an average determination coefficient of R² = 0.9582 with substantially reduced prediction errors (RMSE = 4.1121, MAE = 3.1786, MAPE = 3.1061). Both qualitative and quantitative analyses confirm that CNN–GRU not only provides stable predictions across multiple monitoring points but also effectively captures sudden deformation fluctuations. These results underscore the potential of the proposed AI-assisted framework as a robust and reliable tool for intelligent monitoring, safety assessment, and early warning in large-scale hydropower facilities. Full article

Drive-through establishments are becoming increasingly prominent in urban transport systems; however, their impacts on traffic generation, spatial form, and sustainability remain insufficiently understood. Conventional trip generation manuals often rely on static predictors, such as gross floor area, which can misrepresent demand in high-turnover, convenience-driven contexts and fail to capture operational, behavioral, and environmental effects. This knowledge gap underscores the need for an integrated framework that supports both effective planning and congestion mitigation, particularly in cities experiencing rapid motorization and shifting mobility behaviors. This study investigated the evolving dynamics in trip generation associated with drive-through services and their influence on urban development patterns. A mixed-methods approach was employed, combining a systematic literature review, meta-analysis of queue data, cross-comparison of trip generation rates from international and Asian datasets, and case-based scenario modeling. The results revealed that drive-throughs intensify high-frequency, impulse-driven vehicle trips, thereby causing congestion, reducing pedestrian accessibility, and reinforcing auto-centric land use configurations, while also enhancing consumer convenience and commercial efficiency. This study contributes to the literature by synthesizing inconsistencies in regional datasets; introducing a systems-based framework that integrates structural, behavioral, and environmental determinants with road network topology; and outlining policy applications that align trip generation with zoning, design standards, and sustainable infrastructure planning. Full article

Securing Industrial Control Systems (ICSs) is critical, but it is made challenging by the constant evolution of cyber threats and the scarcity of labeled attack data in these specialized environments. Standard intrusion detection systems (IDSs) often fail to adapt when transferred to new networks with limited data. To address this, this paper introduces an adaptive intrusion detection framework that combines a hybrid Convolutional Neural Network and Long Short-Term Memory (CNN-LSTM) model with a novel transfer learning strategy. We employ a Reinforcement Learning (RL) agent to intelligently guide the fine-tuning process, which allows the IDS to dynamically adjust its parameters such as layer freezing and learning rates in real-time based on performance feedback. We evaluated our system in a realistic data-scarce scenario using only 50 labeled training samples. Our RL-Guided model achieved a final F1-score of 0.9825, significantly outperforming a standard neural fine-tuning model (0.861) and a target baseline model (0.759). Analysis of the RL agent’s behavior confirmed that it learned a balanced and effective policy for adapting the model to the target domain. We conclude that the proposed RL-guided approach creates a highly accurate and adaptive IDS that overcomes the limitations of static transfer learning methods. This dynamic fine-tuning strategy is a powerful and promising direction for building resilient cybersecurity defenses for critical infrastructure. Full article

The macro-economic growth rate of Peru is analyzed for sensitivity to climatic conditions. Year-on-year fluctuations in the inflation-adjusted gross domestic product (GDP) per capita over the period 1970–2024 are subjected to correlation and composite statistical methods. Upturns relate to cool east Pacific La Niña, downturns with warm El Niño. Composites are analyzed by subtracting upper and lower terciles, representing a difference of ~USD 40 B at current value. These reveal how the regional climate exerts a partial influence among external factors. During the austral summer with southeasterly winds over the east Pacific, sea temperatures undergo a 2.5 °C cooling. Consequently, atmospheric subsidence draws humid air from the Amazon toward the Peruvian highlands, improving crop production. Dry weather along the coast sustains transportation networks and urban infrastructure, ensuring good economic performance over the year. The opposing influence of El Niño is built into the statistics. A multi-variate algorithm is developed to predict changes in the Peru growth rate. Austral summer winds and subsurface temperatures over the tropical east Pacific account for a modest 23% of year-on-year variance. Although external factors and the varied landscape weaken macro-economic links with climate, our predictors significantly improve on traditional indices: SOI and Nino3. Adaptive measures are suggested to take advantage of Southern Oscillation’s influence on Peru’s economy. Full article

Cloudbursts and other extreme rainfall events are becoming more frequent and intense, making precise forecasts and disaster preparedness more challenging. Despite advances in meteorological monitoring, current models often lack the precision needed for hyperlocal extreme rainfall forecasts. This study addresses the research gap in spatial configuration-aware modeling by proposing a novel framework that combines geometry-based weather station selection with advanced deep learning architectures. The primary goal is to utilize real-time data from well-placed Automatic Weather Stations to enhance the precision and reliability of extreme rainfall predictions. Twelve unique datasets were generated using four different geometric topologies—linear, triangular, quadrilateral, and circular—centered around the target station Chinchwad in Pune, India, a site that has recorded diverse rainfall intensities, including a cloudburst event. Using common performance criteria, six deep learning models were trained and assessed across these topologies. The proposed Bi-GRU model under linear topology achieved the highest predictive accuracy (R² = 0.9548, RMSE = 2.2120), outperforming other configurations. These findings underscore the significance of geometric topology in rainfall prediction and provide practical guidance for refining AWS network design in data-sparse regions. In contrast, the Transformer model showed poor generalization with high MAPE values. These results highlight the critical role of spatial station configuration and model architecture in improving prediction accuracy. The proposed framework enables real-time, location-specific early warning systems capable of issuing alerts 2 h before extreme rainfall events. Timely and reliable predictions support disaster risk reduction, infrastructure resilience, and community preparedness, which are essential for safeguarding lives and property in vulnerable regions. Full article

The inspection and maintenance of urban sewer infrastructure remain critical challenges for megacities, where conventional manual inspection approaches are labor-intensive, time-consuming, and prone to human error. Although deep learning has been increasingly applied to sewer inspection, the field lacks both a publicly available large-scale dataset and a systematic evaluation of CNN and transformer-based models on real sewer footage. The primary aim of this study is to systematically evaluate and compare state-of-the-art deep learning architectures for automated sewer defect detection using a newly introduced dataset. We present the Istanbul Sewer Defect Dataset (ISWDS), comprising 13,491 expert-annotated images collected from Istanbul’s wastewater network and covering eight defect categories that account for approximately 90% of reported failures. The scientific novelty of this work lies in both the introduction of the ISWDS and the first systematic benchmarking of YOLO (v8/11/12) and RT-DETR (v1/v2) architectures under identical protocols on real sewer inspection footage. Experimental results demonstrate that RT-DETR v2 achieves the best performance (F1: 79.03%, Recall: 81.10%), significantly outperforming the best YOLO variant. While transformer-based architectures excel in detecting partially occluded defects and complex operational conditions, YOLO models provide computational efficiency advantages for resource-constrained deployments. Furthermore, a QGIS-based inspection tool integrating the best-performing models was developed to enable real-time video analysis and automated reporting. Overall, this study highlights the trade-offs between accuracy and efficiency, demonstrating that RT-DETR v2 is most suitable for server-based processing. In contrast, compact YOLO variants are more appropriate for edge deployment. Full article