Search Results (1,330)

Rapid data growth in large systems has introduced significant challenges in real-time monitoring and analysis. One of these challenges is detecting anomalies in time series data with high-dimensional inputs that contain complex inter-correlations between them. In addition, the lack of labeled data leads to the use of unsupervised learning that relies on daily system data to train models, which can contain noise that affects feature extraction. To address these challenges, we propose PCA-AAE, a novel anomaly detection model for time series data using an Adversarial Autoencoder integrated with Principal Component Analysis (PCA). PCA contributes to analyzing the latent space by transforming it into uncorrelated components to extract important features and reduce noise within the latent space. We tested the integration of PCA into the model’s phases and studied its efficiency in each phase. The tests show that the best practice is to apply PCA to the latent code during the adversarial training phase of the AAE model. We used two public datasets, the SWaT and SMAP datasets, to compare our model with state-of-the-art models. The results indicate that our model achieves an average F1 score of 0.90, which is competitive with state-of-the-art models, and an average of 58.5% faster detection speed compared to similar state-of-the-art models. This makes PCA-AAE a candidate solution to enhance real-time anomaly detection in high-dimensional datasets. Full article

Real-time traffic light detection is essential for the safe navigation of autonomous vehicles, where timely and accurate recognition of signal states is critical. YOLOv8, a state-of-the-art object detection model, offers enhanced speed and precision, making it well-suited for real-time applications in complex driving environments. This study presents a modified YOLOv8 architecture optimized for traffic light detection by integrating Depth-Wise Separable Convolutions (DWSCs) throughout the backbone and head. The model was first pretrained on a public traffic light dataset to establish a strong baseline and then fine-tuned on a custom real-time dataset consisting of 480 images collected from video recordings under diverse road conditions. Experimental results demonstrate high detection performance, with precision scores of 0.992 for red, 0.995 for yellow, and 0.853 for green lights. The model achieved an average mAP@0.5 of 0.947, with stable F1 scores and low validation losses over 80 epochs, confirming effective learning and generalization. Compared to existing YOLO variants, the modified architecture showed superior performance, especially for red and yellow lights. Full article

Background and Objectives: Dual-task training (DTT) is an innovative therapeutic approach that involves the simultaneous application of two tasks, which can be motor, cognitive, or a combination of both. Children with cerebral palsy (CP) often exhibit impairments in balance, motor skills, and gait, conditions that may be amenable to improvement through DTT. The aim of this study was to determine the effectiveness of DTT in enhancing balance, walking speed, and gross motor function-related balance in children with CP. Materials and Methods: In accordance with PRISMA guidelines, a comprehensive systematic review with meta-analysis (SRMA) was conducted. Electronic databases like PubMed Medline, Scopus, Web of Science, CINAHL, and PEDro were searched up to March 2025, with no language or publication date restrictions. Only randomized controlled trials (RCTs) examining the effectiveness of DTT on balance, gross motor function, and walking speed in children with CP were included. The methodological quality and risk of bias of the included RCTs were assessed using the PEDro scale. Pooled effects were calculated using Cohen’s standardized mean difference (SMD) and its 95% confidence interval (95% CI) within random-effects models. Results: Eight RCTs, providing data from 216 children, were included. Meta-analyses suggested that DTT was more effective than conventional therapies for increasing functional (SMD = 0.65; 95% CI 0.18 to 1.13), dynamic (SMD = 0.61; 95% CI 0.15 to 1.1), and static balance (SMD = 0.46; 95% CI 0.02 to 0.9), as well as standing (SMD = 0.75; 95% CI 0.31 to 1.18; p = 0.001) and locomotion dimensions (SMD = 0.65; 95% CI 0.22 to 1.08) of the Gross Motor Function Measure (GMFM) and walking speed (SMD = 0.46; 95% CI 0.06 to 0.87). Subgroup analyses revealed that a motor–cognitive dual task is better than a motor single task for functional, dynamic, and static balance and standing and locomotion dimensions for the GMFM. Conclusions: This SRMA, including the major number of RCTs to date, suggests that DTT is effective in increasing balance, walking and gross motor function-related balance in children with CP. Full article

GPSs typically have an accuracy ranging from a few meters to several tens of meters. However, when corrected using various methods, they can achieve an accuracy of several tens of centimeters. In autonomous driving, a positioning accuracy of less than 50 cm is required for lane-level positioning, route generation, and navigation. However, in environments where GPS signals are blocked, such as tunnels and underground roads, absolute positioning is impossible. Instead, relative positioning methods integrating IMU, IVN, and cameras are used. These methods are influenced by numerous variables, however, such as vehicle speed and road conditions, resulting in lower accuracy. In this study, we conducted experiments on current vehicle navigation technologies using an autonomous driving simulation vehicle in the Suri–Suam Tunnel of the Seoul Metropolitan Area 1st Ring Expressway. To recognize objects (lane markings/2D/3D) for position correction inside the tunnel, data on tunnel and underground road infrastructure in Seoul and Gyeonggi Province was collected, processed, refined, and trained. Additionally, a Loosely Coupled-based Kalman Filter was designed and applied for the fusion of GPSs, IMUs, and IVNs. As a result, an error of 113.62 cm was observed in certain sections. This suggests that while the technology is applicable for general vehicle lane-level navigation in ultra-long tunnels spanning several kilometers for public service, it falls short of meeting the precision required for autonomous driving systems, which demand lane-level accuracy. Therefore, it was concluded that infrastructure-based absolute positioning technology is necessary to enable precise navigation inside tunnels. Full article

The Union of the Comoros, located in the Indian Ocean, faces persistent energy challenges due to its geographic isolation, heavy dependence on imported fossil fuels, and underdeveloped electricity infrastructure. This study investigates the techno-economic optimization of a hybrid microgrid designed to supply electricity to a rural village in Grande Comore. The proposed system integrates photovoltaic (PV) panels, wind turbines, a diesel generator, and battery storage. Detailed modeling and simulation were conducted using HOMER Energy, accompanied by a sensitivity analysis on solar irradiance, wind speed, and diesel price. The results indicate that the optimal configuration consists solely of PV and battery storage, meeting 100% of the annual electricity demand with a competitive levelized cost of energy (LCOE) of 0.563 USD/kWh and zero greenhouse gas emissions. Solar PV contributes over 99% of the total energy production, while wind and diesel components remain unused under optimal conditions. Furthermore, the system generates a substantial energy surplus of 63.7%, which could be leveraged for community applications such as water pumping, public lighting, or future system expansion. This study highlights the technical viability, economic competitiveness, and environmental sustainability of 100% solar microgrids for non-interconnected island territories. The approach provides a practical and replicable decision-support framework for decentralized energy planning in remote and vulnerable regions. Full article

Personal Rapid Transit (PRT) and Group Rapid Transit (GRT) are classes of fully automated public transport systems, where passengers can travel in small vehicles on an interconnected, grade-separated network of guideways, non-stop, from origin to destination. PRT and GRT are considered sustainable as they are low-emission and able to attract car drivers. The parameterized cost modeling framework developed in this paper has the advantage that profitability of different PRT/GRT systems can be rapidly verified in a transparent way and in function of a variety of relevant system parameters. This framework may contribute to a more transparent, rapid, and low-cost evaluation of PRT/GRT schemes for planning and decision-making purposes. The main innovation is the introduction of the “peak hour network saturation” S: the number of vehicles in circulation during peak hour divided by the maximum number of vehicles running at line speed with minimum time headways. It is an index that aggregates the main uncertainties in the planning process, namely the demand level relative to the supply level. Furthermore, a maximum S can be estimated for a PRT/GRT project, even without a detailed demand estimation. The profit per trip is analytically derived based on S and a series of more certain parameters, such as fares, capital and maintenance costs, daily demand curve, empty vehicle share, and physical properties of the system. To demonstrate the ability of the framework to analyze profitability in function of various parameters, we apply the methods to a single vehicle PRT, a platooned PRT, and a mixed PRT/GRT. The results show that PRT services with trip length proportional fares could be profitable already for $S>0.25$. The PRT capacity, profitability, and robustness to tripled infrastructure costs can be increased by vehicle platooning or GRT service during peak hours. Full article

The typical ranges available for electric vehicles (EVs) may be considered by users to be inadequate when compared to long, real-life trips, and charging operations may need to be planned along journeys. To evaluate the compatibility between vehicle features and charging options for realistic journeys performed by car, a simulation approach is proposed here, using travel data collected from real vehicles to obtain trip chains for multiple consecutive days. Car travel activities, including stops with the option of charging, were simulated by applying an agent-based approach. Charging operations can be integrated into trip chains for user activities, assuming that they remain unchanged in the event that vehicles switch to electric. The energy consumption of the analyzed trips, disaggregated by vehicle type, was estimated using the average travel speed, which is useful for capturing the main route features (ranging from urban to motorways). Data were recorded for approximately 25,000 vehicles in the Turin Metropolitan Area for six consecutive days. Market segmentation of the vehicles was introduced to take into consideration different energy consumption rates and charging times, given that the electric power, battery size, and consumption rate can be related to the vehicle category. Charging activities carried out using public infrastructure during idle time between consecutive trips, as well as those carried out at home or work, were identified in order to model different needs. Full article

Background/Objectives: This study aimed to evaluate sociodemographic factors, features of the acute infection, and post-infection health status in survivors of COVID-19, assessing their association with post-acute COVID-19 syndrome (PACS). Methods: A multidisciplinary public clinic in Brazil assessed COVID-19 survivors between June 2020 and February 2022. Patients were classified as having PACS or subacute infection (SI). Data on the history of the acute infection, current symptoms, physical examination, and laboratory findings were collected and analyzed using multivariate models with PACS as the outcome. Results: Among the 113 participants, 63.71% were diagnosed with PACS at a median of 130 days (IQR: 53–196) following acute symptom onset. Admission to the intensive care unit was more frequent among individuals with PACS than those with SI (83.3% vs. 65.0% respectively; p = 0.037). Symptoms significantly more prevalent in the PACS group when compared to the SI cohort included hair loss (44.4% vs. 17.1% respectively; p = 0.004), lower limb paresthesia (34.7% vs. 9.8% respectively; p = 0.003), and slow thinking speed (28.2% vs. 0.0% respectively; p < 0.001). Logistic regression revealed that only the time interval between the onset of acute symptoms and the clinical evaluation was independently associated with a PACS diagnosis (β = 0.057; 95% CI: 1.03–1.08; p < 0.001). Conclusions: Patients with PACS had a higher frequency of intensive care unit admission compared to those with subacute infection. However, in the multivariate analysis, the severity of the acute infection did not predict the final diagnosis of PACS, which was associated only with the time elapsed since symptom onset. Full article

Background: Retinal diseases are becoming an important public health issue, with early diagnosis and timely intervention playing a key role in preventing vision loss. Optical coherence tomography (OCT) remains the leading non-invasive imaging technique for identifying retinal conditions. However, distinguishing between diabetic macular edema (DME) and macular edema resulting from retinal vein occlusion (RVO) can be particularly challenging, especially for clinicians without specialized training in retinal disorders, as both conditions manifest through increased retinal thickness. Due to the limited research exploring the application of deep learning methods, particularly for RVO detection using OCT scans, this study proposes a novel diagnostic approach based on stacked convolutional neural networks. This architecture aims to enhance classification accuracy by integrating multiple neural network layers, enabling more robust feature extraction and improved differentiation between retinal pathologies. Methods: The VGG-16, VGG-19, and ResNet50 models were fine-tuned using the Kermany dataset to classify the OCT images and afterwards were trained using a private OCT dataset. Four stacked models were then developed using these models: a model using the VGG-16 and VGG-19 networks, a model using the VGG-16 and ResNet50 networks, a model using the VGG-19 and ResNet50 models, and finally a model using all three networks. The performance metrics of the model includes accuracy, precision, recall, $F_2$-score, and area under of the receiver operating characteristic curve (AUROC). Results: The stacked neural network using all three models achieved the best results, having an accuracy of 90.7%, precision of 99.2%, a recall of 90.7%, and an $F_2$-score of 92.3%. Conclusions: This study presents a novel method for distinguishing retinal disease by using stacked neural networks. This research aims to provide a reliable tool for ophthalmologists to improve diagnosis accuracy and speed. Full article

Unmanned aerial vehicles (UAVs) are increasingly used for road infrastructure inspection and monitoring. However, challenges such as scale variation, complex background interference, and the scarcity of annotated UAV datasets limit the performance of traditional segmentation models. To address these challenges, this study proposes UAV-YOLOv12, a multi-scale segmentation model specifically designed for UAV-based road imagery analysis. The proposed model builds on the YOLOv12 architecture by adding two key modules. It uses a Selective Kernel Network (SKNet) to adjust receptive fields dynamically and a Partial Convolution (PConv) module to improve spatial focus and robustness in occluded regions. These enhancements help the model better detect small and irregular road features in complex aerial scenes. Experimental results on a custom UAV dataset collected from national highways in Wuxi, China, show that UAV-YOLOv12 achieves F1-scores of 0.902 for highways (road-H) and 0.825 for paths (road-P), outperforming the original YOLOv12 by 5% and 3.2%, respectively. Inference speed is maintained at 11.1 ms per image, supporting near real-time performance. Moreover, comparative evaluations with U-Net show that UAV-YOLOv12 improves by 7.1% and 9.5%. The model also exhibits strong generalization ability, achieving F1-scores above 0.87 on public datasets such as VHR-10 and the Drone Vehicle dataset. These results demonstrate that the proposed UAV-YOLOv12 can achieve high accuracy and robustness in diverse road environments and object scales. Full article

Rapid industrialization and urbanization have progressed in Korea, yet public attention to hazardous pollutants emitted from industrial complexes remains limited. With the increasing coexistence of industrial and residential areas, there is a growing need for real-time monitoring and management plans that account for the rapid dispersion of hazardous air pollutants (HAPs). In this study, we conducted spatiotemporal data collection and analysis for the first time in Korea using real-time measurements obtained through mobile extractive differential optical absorption spectroscopy (Me-DOAS) mounted on a solar occultation flux (SOF) vehicle. The measurements were conducted in the Saha Sinpyeong–Janglim Industrial Complex in Busan, which comprises the Sasang Industrial Complex and the Sinpyeong–Janglim Industrial Complex. BTEX compounds were selected as target volatile organic compounds (VOCs), and real-time measurements of both BTEX and fine particulate matter (PM) were conducted simultaneously. Correlation analysis revealed a strong relationship between PM₁₀ and PM_2.5 (r = 0.848–0.894), indicating shared sources. In Sasang, BTEX levels were associated with traffic and localized facilities, while in Saha Sinpyeong–Janglim, the concentrations were more influenced by industrial zoning and wind patterns. Notably, inter-compound correlations such as benzene–m-xylene and p-xylene–toluene suggested possible co-emission sources. This study proposes a GIS-based, three-dimensional air quality management approach that integrates variables such as traffic volume, wind direction, and speed through real-time measurements. The findings are expected to inform effective pollution control strategies and future environmental management plans for industrial complexes. Full article

The Latin American and Caribbean (LAC) region faces persistent challenges of inequality, climate change vulnerability, and deteriorating air quality. The Aburrá Valley, where Medellín is located, is a narrow tropical valley with complex topography, strong thermal inversions, and unstable atmospheric conditions, all of which exacerbate the accumulation of pollutants. In Medellín, ${\mathrm{NO}}_2$ concentrations have remained nearly unchanged over the past eight years, consistently approaching critical thresholds, despite the implementation of air quality control strategies. These persistent high concentrations are closely linked to the variability of the atmospheric boundary layer (ABL) and are often intensified by prolonged dry periods. This study focuses on a representative street canyon in Medellín that has undergone recent urban interventions, including the construction of new public spaces and pedestrian areas, without explicitly considering their impact on ${\mathrm{NO}}_x$ dispersion. Using Computational Fluid Dynamics (CFD) simulations, this work evaluates the influence of urban morphology on ${\mathrm{NO}}_x$ accumulation. The results reveal that areas with high Aspect Ratios (AR > 0.65) and dense vegetation exhibit reduced wind speeds at the pedestrian level—up to 40% lower compared to open zones—and higher ${\mathrm{NO}}_2$ concentrations, with maximum simulated values exceeding 50 μg/m³. This study demonstrates that the design of pedestrian corridors in complex urban environments like Medellín can unintentionally create pollutant accumulation zones, underscoring the importance of integrating air quality considerations into urban planning. The findings provide actionable insights for policymakers, emphasizing the need for comprehensive modeling and field validation to ensure healthier urban spaces in cities affected by persistent air quality issues. Full article

Large-scale events, such as festivals and public gatherings, pose serious problems in terms of traffic congestion, slow transaction processing, and security risks to transportation planning. This study proposes a blockchain-based solution for enhancing the efficiency and security of intelligent transport systems (ITS) by utilizing state channels and rollups. Throughput is optimized, enabling transaction speeds of 800 to 3500 transactions per second (TPS) and delays of 5 to 1.5 s. Prevent data tampering, strengthen security, and enhance data integrity from 89% to 99.999%, as well as encryption efficacy from 90% to 98%. Furthermore, our system reduces congestion, optimizes vehicle movement, and shares real-time, secure data with stakeholders. Practical applications include fast and safe road toll payments, faster public transit ticketing, improved emergency response coordination, and enhanced urban mobility. The decentralized blockchain helps maintain trust among users, transportation authorities, and event organizers. Our approach extends beyond large-scale events and proposes a path toward ubiquitous, Artificial Intelligence (AI)-driven decision-making in a broader urban transit network, informing future operations in dynamic traffic optimization. This study demonstrates the potential of blockchain to create more intelligent, more secure, and scalable transportation systems, which will help reduce urban mobility inefficiencies and contribute to the development of resilient smart cities. Full article

The Yangtze River Delta urban agglomeration, characterized by high population density, an advanced transportation system, and a concentration of industrial activity, is one of the regions severely affected by O₃ pollution in central and eastern China. Using data collected from 251 monitoring stations between 2015 and 2025, this paper analyzed the spatio-temporal variation of 8 h O₃ concentrations and instances of exceedance. On the basis of exploring the influence of meteorological factors on regional 8 h O₃ concentration, the potential source contribution areas of pollutants under the exceedance condition were investigated using the HYSPLIT model. The results indicate a rapid increase in the 8 h O₃ concentration at a rate of 0.91 ± 0.98 μg·m⁻³·a⁻¹, with the average number of days exceeding concentration standards reaching 41.05 in the Yangtze River Delta urban agglomeration. Spatially, the 8 h O₃ concentrations were higher in coastal areas and lower in inland regions, as well as elevated in plains compared to hilly terrains. This distribution was significantly distinct from the concentration growth trend characterized by higher levels in the northwest and lower levels in the southeast. Furthermore, it diverged from the spatial characteristics where exceedances primarily occurred in the heavily industrialized northeastern region and the lightly industrialized central region, indicating that the growth and exceedance of 8 h O₃ concentrations were influenced by disparate factors. Local human activities have intensified the emissions of ozone precursor substances, which could be the key driving factor for the significant increase in regional 8 h O₃ concentrations. In the context of high temperatures and low humidity, this has contributed to elevated levels of 8 h O₃ concentrations. When wind speeds were below 2.5 m·s⁻¹, the proportion of 8 h O₃ concentrations exceeding the standards was nearly 0 under almost calm wind conditions, and it showed an increasing trend with rising wind speeds, indicating that the potential precursor sources that caused high O₃ concentrations originated occasionally from inland regions, with very limited presence within the study area. This observation implies that the main cause of exceedances was the transport effect of pollution from outside the region. Therefore, it is recommended that the Yangtze River Delta urban agglomeration adopt economic and technological compensation mechanisms within and between regions to reduce the emission intensity of precursor substances in potential source areas, thereby effectively controlling O₃ concentrations and improving public living conditions and quality of life. Full article

Electric buses are key in the strategy towards a greenhouse-gas-neutral fleet. However, their restrictions in terms of range and refueling as well as their increased price point present new challenges for public transport companies. This study aims to address, based on real-world operational data, how energy consumption and charging behavior affect battery aging and how operational strategies can be optimized to extend battery life under realistic conditions. This article presents an energy consumption analysis with respect to ambient temperatures and average vehicle speed based exclusively on real-world data of an urban bus fleet, providing a data foundation for range forecasting and infrastructure planning optimized for public transport needs. Additionally, the State of Charge (SOC) window during operation and vehicle idle time as well as the charging power were analyzed in this case study to formulate recommendations towards a more battery-friendly treatment. The central research question is whether battery-friendly operational strategies—such as reduced charging power and lower SOC windows—can realistically be implemented in daily public transport operations. The impact of the recommendations on battery lifetime is estimated using a battery aging model on drive cycles. Finally, the reduction in CO₂ emissions compared to diesel buses is estimated. Full article