Reliable long-range onboard perception is a prerequisite for future railway safety systems, where potential obstacles must be recognized under long braking distances, sparse far-field returns, and strongly constrained rail-corridor geometry. This paper presents Rail-BEV as an initial reproducible baseline study for LiDAR-centric, sensor-aware bird's-eye-view (BEV) railway obstacle perception. LiDAR is used as the primary geometric sensing modality, while a front-center RGB camera provides lightweight auxiliary visual evidence through calibrated LiDAR-to-image projection. The aligned geometric and visual cues are organized within a unified railway-oriented BEV backend that integrates geometry-aware fusion, rail-geometry prediction, and lightweight inference-time structural refinement. Evaluation was conducted on a scene-isolated railway benchmark with range-stratified center-distance matching, and all model variants were assessed on independent test sequences rather than on validation-selected checkpoints. Compared with CenterPoint and BEVFusion baselines evaluated under the same settings, Rail-BEV achieved the highest overall mAP of 0.6669, with particularly improved long-range pedestrian perception. The controlled ablation further shows that front-view RGB evidence improves the LiDAR-only baseline from 0.5612 to 0.5750 mAP, while ROI-based rail-corridor refinement further increases mAP to 0.5916 and Rail-BEV mIoU to 0.1193. These results indicate that LiDAR-centered sensing, lightweight visual assistance, and coarse rail-aware structural reasoning can be jointly organized to support reproducible long-range railway obstacle perception. This study also clarifies the remaining limitations in rail-geometry quality, calibration robustness, sensor degradation, and strict railway-oriented localization.
The material heterogeneity and structural discontinuity of welded rail joints in high-speed railways make them vulnerable to fatigue damage under wheel-rail rolling contact. In this study, a three-dimensional finite element model of a 1380 MPa carbide-free bainitic rail welded joint was developed, and the fatigue damage behavior was assessed using the Findley critical-plane multiaxial fatigue criterion. The results show that the welded joint causes pronounced stress concentration, with the peak contact stress increasing from approximately 1000 MPa in the base metal to about 1120 MPa near the weld center. Fatigue damage is mainly concentrated in the weld zone and gradually decreases along the longitudinal direction toward both sides. The maximum fatigue damage occurs in the subsurface region approximately 2-3 mm beneath the rail surface. As the wheel load increases from 80 kN to 120 kN, the fatigue damage in the weld zone increases markedly. A clear coupling effect between wheel load and friction coefficient is also observed, indicating that higher wheel loads and friction coefficients accelerate fatigue damage accumulation. These findings identify the weld zone as the most fatigue-critical region under rolling contact loading and indicate that the subsurface layer at a depth of 2-3 mm is the preferential site for fatigue crack initiation. The results provide a useful reference for life assessment and maintenance management of welded rail joints in high-speed railways.
Environmental exposures, such as railway noise and vibration, might be unequally distributed across communities. This study investigates how area-level socioeconomic conditions and migration density vary with the distribution of railway noise and vibration in Southwest Sweden. We employed linear regression and spatial autoregressive models to analyse data from the EpiVib study, which included 7,280 individuals living within 1 km of a trafficked railway in four regions of Southwest Sweden. The study assessed the relationships between exposure levels and area-level socioeconomic conditions and migration density using Statistics Sweden's small-area division system (i.e., DeSO). Areas with lower socioeconomic conditions and higher migration density are exposed to higher levels of railway noise and vibration. Spatial autocorrelation analysis revealed that noise levels were higher in areas with more than 20% of inhabitants having a foreign background. The impact of area-level indicators on noise exposure varied with urbanicity, with higher exposure levels in rural and suburban areas compared to urban areas. Vibration exposure was more pronounced in areas with fair socioeconomic conditions and higher migration density. The study shows inequalities in railway noise and vibration exposure, according to area-level socioeconomic conditions and migration density. Areas with better socioeconomic conditions are consistently less exposed to both railway noise and vibration. Furthermore, the impact of area-level indicators on noise exposure depends on urbanicity levels. Future research, interventions and policies that account for area-level socioeconomic factors as well as urbanicity are advised to ensure environmental equity and justice regarding rail traffic exposures.
Linear infrastructures are rapidly expanding worldwide, however, their ecological impacts remain poorly understood. Here we integrate vegetation, soil, and microbial surveys along a 495-km north-south transect of the recently opened China-Laos Railway to assess how railway construction influences above-belowground interactions and plant invasion dynamics, and how this depends on latitude and climate. Station habitats were characterized by lower soil organic carbon, total nitrogen, and moisture, as well as higher pH and altered C:N:P stoichiometry. These abiotic changes were accompanied by a pronounced reorganization of soil fungal communities, manifested as reduced beta diversity and increased richness of inferred plant pathogenic and arbuscular mycorrhizal fungal guilds. Disturbed vegetation near railway stations supported significantly greater alien plant richness, cover, and biomass compared to adjacent undisturbed vegetation. Moreover, the mean phylogenetic distance between alien and native species decreased with increasing latitude and remained higher in less disturbed sites, suggesting a transition from species interaction-driven exclusion in the tropics to environment-driven filtering in temperate regions. Structural equation modeling suggested that railway construction was associated with higher alien plant richness through multiple pathways, including shifts in soil nutrient stoichiometry, reduced native plant richness, and increased abundance of potential plant pathogenic fungi. Native plant richness exhibited the strongest negative relationship with alien plant richness. These findings suggest that railway-associated changes in environmental conditions and native community structure may influence patterns of plant invasion along railway corridors.
Rail surface defect detection faces significant challenges, including class imbalance, small-scale defect detection, and interference from complex backgrounds. To address these issues, this study proposes a lightweight rail surface defect detection method based on YOLOv8n. First, a K-means +  + -based defect scale prior optimization strategy is introduced to better characterize the scale distribution of rail surface defects and improve the representation of small targets. Second, a local-global collaborative attention module, named CloAtt, is embedded into the C2f. structure to enhance fine-grained defect feature representation while suppressing complex background noise. Third, an Attention-Guided Bidirectional Feature Pyramid Network, named ABiFPN, is designed to improve multi-scale feature fusion and reduce redundant feature propagation. Furthermore, a Focal-Efficient Intersection over Union loss, named Focal-EIoU, is adopted to improve bounding box regression and enhance the learning of difficult samples. Experimental results on the ProRail dataset show that the proposed YOLOv8n-ours achieves an mAP@0.5 of 85.7%, with only 2.13M parameters and 7.5 GFLOPs. Compared with the baseline YOLOv8n, the proposed method improves detection accuracy while reducing model complexity. These results demonstrate that the proposed model provides an effective lightweight solution for rail surface defect detection. Edge-device deployment and real-time inference validation will be investigated in future work.
In high-risk transport environments, incidents are often linked with critical issues in team communication and decision-making. One factor receiving growing attention in rail transportation is the phenomena of 'authority gradients'-imbalances in perceived power or status that can inhibit speaking up, suppress dissent, and compromise safety. While authority gradients are widely acknowledged in other industries, their role in rail remains poorly articulated and subsumed under vague terminology. This lack of specificity can hinder accurate diagnosis of risks and limit the effectiveness of safety interventions. The aim of this study was to examine how authority gradients are identified, described, and conceptualised in real-world rail incident investigations. A total of 786 reports published by formal accident investigation agencies in five countries (Australia, UK, Canada, USA, New Zealand) over a 10-year period were acquired. Following screening, selection, and data extraction based on PRISMA, a review and meta-synthesis of 48 incidents was conducted using a typology of authority gradient presentation based on four classes of presentation. In total, 12 specific behaviours synonymous with authority gradients were found, with failure to challenge (n = 19) and conformity/pressure to comply (n = 17) the most prevalent. Despite this, only one report attributed an 'authority gradient,' while 7 made overt references but did not identify it as a causal or contributory factor. Twenty-one reports implicated authority gradients, and 19 were suggestive of power relationships with potential to influence error. This research elucidates how rail jurisdictions consider and attribute authority gradients, and underscores a need for increased focus on the relational aspects of team interaction.
Previous research has reported a significant effect of refractive defocus on the correct identification of red signals. The purpose of this study is to investigate the effects of both refractive defocus and non-refractive defocus (using Bangerter filters) on the perception of rail signals using the Railway LED Lantern Test (RLLT). The RLLT is the simulated practical test nominated in the Australian National Standard for Health Assessment of Railway Safety Workers. Participants were 19-59 years old, and the best corrected visual acuity (BVCA) was required to be no worse than 6/9 binocularly. Subjects with current or active ocular conditions were excluded, and sufficiency in English was required. Best corrected refraction, visual acuity and colour vision were assessed. Participants carried out the RLLT binocularly under five conditions: best corrected, +0.50 DS, +0.75 DS and Bangerter filters 1.0 and 0.8. Ten male and 10 female subjects completed the study; age range 20-25 and mean age 22.4 ± 1.1 years. BVCA was 6/6 (logMAR 0.0 or better). Errors occurred far more often with red than with yellow or green (p < 0.0001) and with Bangerter filter blur more than refractive blur (p < 0.0001). Failing to see a red signal, rather than misnaming the red as yellow or green, was the predominant error (p < 0.0001) and was induced far more frequently by Bangerter filters than refractive blur (p < 0.0001). This error was far more common than miscalling red as yellow (p < 0.0001) (paired t-tests). These findings suggest that a large proportion of errors are due to not seeing the red signal rather than miscalling the red as yellow or green. Non-refractive blur was found to cause a greater increase in colour errors.
Indirect, onboard monitoring of railway track infrastructure is critical for improving safety and operational efficiency, yet its adoption remains limited due to the complexity of real-world data. Development has historically followed two separate paths: vibration analysis, which offers high sensitivity to subsurface anomalies but lacks spatial context, and vision-based inspection, which provides spatial context but is constrained by environmental conditions and limited to surface-level detection. Progress in both approaches-especially their integration-has been limited by the lack of a publicly available benchmark dataset for controlled-condition analysis, including multimodal cases. To address this gap, we present a novel temporally synchronized railroad vibration and vision dataset (Rail-VIVID). This dataset, collected from repeated runs over a track segment with documented anomalies under controlled conditions, serves two main goals. It enables investigation into the relationship between track anomalies and their signatures in each sensing modality by serving as a benchmark for anomaly detection algorithms. It also provides a foundation for developing integrated, multimodal sensor fusion approaches to support more robust detection and predictive maintenance.
Personalization of musculoskeletal (MSK) spine models requires subject-specific paravertebral muscle morphology, which is commonly obtained using magnetic resonance imaging (MRI) or computed tomography (CT). However, these modalities are costly, time-intensive, and, in the case of CT, involve ionizing radiation. The purpose of this study was to develop a custom-designed rail-guided ultrasound system and evaluate its intra- and inter-rater reliability for standardized assessment of thoracolumbar paravertebral muscle morphology. The developed height-adjustable system guides an ultrasound probe along three orthogonal axes and allows controlled rotation and application of a constant inward force to minimize operator-dependent variability. Twelve healthy adults participated in two measurement sessions within one week. Cross-sectional area (CSA) and muscle thickness (MT) were measured bilaterally at spinal levels L4, L1, and T8 using a curved-array transducer. Intra-rater reliability was assessed across sessions (n= 12), and inter-rater reliability was evaluated in a subset of five participants in the first session. All measurements were analyzed using intraclass correlation coefficients (ICCs), Bland-Altman plots, and root mean square error. Intra-rater reliability was excellent for CSA (ICC[1,1] = 0.994) and good for MT (ICC[1,1] = 0.882). Inter-rater reliability was excellent for CSA (ICC[3,1] = 0.952) and good for MT (ICC[3,1] = 0.877). Bland-Altman analysis showed minimal bias and narrow limits of agreement for both parameters. The rail-guided ultrasound system enabled reproducible bilateral measurements across lumbar and thoracic levels. With further validation, this radiation-free, cost-effective, and laboratory-based approach may provide a feasible alternative to MRI or CT for obtaining subject-specific paravertebral muscle morphology and facilitate scalable personalization of MSK spine models.
We present a 3D laser-scanning method for the fast, accurate dimensional inspection of large high-speed-rail precast box girders. The pipeline uses low-pass filtering plus sequential registration to suppress noise, and voxel filtering with curvature-aware enhancement to reduce point cloud size by 3-5× while preserving key geometry. Reconstruction employs K-nearest-neighbors and PCA to detect boundaries and curvature jumps, B-spline fitting with moving least squares for surface completion, and CSS corner detection to extract key dimensions at millimeter precision. Field tests report absolute errors ≤ 2.0 mm versus manual measurement, validating the method for automated, digital acceptance.
As urban rail transit networks become increasingly dense, new tunnels frequently undercross existing operational lines. The cyclic loads from existing double-track trains pose a threat to the structural safety of the underlying sections. This paper proposes a computationally efficient analytical model for the 'double-track train-soil layer-underpass tunnel section' system. In this model, the double-track train is represented as a moving two-degree-of-freedom (2-DOF) sprung mass system, the soil is modeled as a spring-damper system, and the underpass tunnel roof structure is treated as an Euler-Bernoulli simply supported beam. The system response is solved using the modal superposition method. Key parameters are calibrated via displacement back analysis, and the model's predictive capability is validated against FLAC3D simulations and independent field measurement data. A systematic parametric study investigates the influence of train speed, tunnel beam length, overburden thickness, equivalent stiffness, and soil mechanical parameters. The analysis reveals the underlying mechanisms of system resonance and critical design parameters. For the specific case study, resonance peaks were observed at speeds around 48 km/h and 72 km/h, a dual-peak resonance phenomenon emerged at beam lengths of approximately 25 m and 45 m, and a vibration amplification effect was identified at a soil cover thickness near 5 m. These findings highlight the resonance mechanisms that should be avoided in design, while the specific critical parameter values are case-dependent and should be re-calibrated for other projects. Increasing the equivalent beam stiffness can effectively suppress the vibration response, while grouting reinforcement modulated the system's dynamic response. This study provides a theoretical framework and an efficient preliminary assessment tool for understanding vibration mechanisms and aiding early-stage design in similar projects.
暂无摘要（点击查看详情）
Ultrasonic guided wave technology offers advantages such as long-distance propagation and full-section detection for non-destructive testing of rails. However, operational rails are typically limited by fastener systems, resulting in increased guided wave energy attenuation and shorter propagation distances compared to free rails. Therefore, studying the mechanism by which fasteners influence guided wave attenuation is crucial. This paper presents a comprehensive three-dimensional solid fastener-rail contact mechanics and transient dynamics coupling model that considers the static action and complex contact relationship between fasteners and rails. The model uses an implicit-explicit combination method to simulate the propagation of guided waves in operational rails. This paper presents a simulation in which a preload force of 20 kN was applied to the fasteners, and the stress distribution of the model was analyzed. A transient analysis of the guided wave was performed under the preload force to investigate the influence of fasteners on guided wave attenuation in rails by comparing attenuation with and without fasteners. Additionally, this paper explores the impact of two contact types, penalty and motion, on wave attenuation in Abaqus/Explicit. The results demonstrate that fasteners significantly exacerbate wave attenuation in rails. The wave packet amplitude calculated using the penalty contact form is slightly greater than that calculated using the motion contact form, and this difference increases with propagation distance.
This focused systematic review and meta-analysis evaluated robot-assisted inguinal lymphadenectomy (RAIL) versus open inguinal lymph node dissection (OILND) in penile cancer, aiming to provide robot-specific comparative estimates rather than a broad minimally invasive synthesis. PubMed, EMBASE, Web of Science, and the Cochrane Library were searched through December 2025 for comparative studies. All meta-analyses were conducted using STATA 18, employing random-effects modeling. Binary outcomes were summarized using odds ratios (ORs), and continuous outcomes were presented as weighted mean differences (WMDs), with all estimates accompanied by 95% confidence intervals (CIs). Risk of bias in the included non-randomized comparative studies was assessed using the ROBINS-I tool. Four retrospective comparative studies involving 276 patients were included, and no randomized controlled trials were identified. A lower odds of total postoperative complications was observed with RAIL (OR = 0.52, 95% CI: 0.27, 0.97; P = 0.047), but this borderline finding should be interpreted cautiously given the small number of retrospective studies and potential residual confounding. Across individual studies, operative time generally tended to be longer with RAIL, whereas findings for estimated blood loss and drainage-related outcomes were inconsistent. Lymph node yield, skin-related complications, minor complications, and groins with positive nodes did not differ significantly between approaches. According to the GRADE assessment, the certainty of evidence was low for most pooled outcomes and very low for skin-related complications and lower-limb edema/lymphedema. RAIL may be associated with lower overall postoperative complications than open surgery, but the evidence remains low certainty and insufficient to establish a definitive morbidity-reduction benefit. In addition, several continuous perioperative variables showed extreme between-study heterogeneity and were not clinically interpretable as reliable pooled effects.
Globally, a significant volume of petroleum products is transported daily through logistic networks to meet diverse regional demands, where unreliable or delayed delivery can lead to serious economic, social, and political consequences. The primary contribution of this research is the development of a comprehensive, robust multi-period mathematical model for the integrated planning of a green multi-modal petroleum product logistic network. This model advances current frameworks by simultaneously integrating pipeline, rail, and road transportation; synchronizing strategic facility development with operational flow allocation under uncertainty; and selecting the optimal network topology from both economic and environmental perspectives. It supports decisions on the location of distribution centers and the construction of pipelines and railways within budget constraints, aligning infrastructure investment with operational efficiency. A real-world case study in central Iran, solved via the Augmented Epsilon Constraint method, validates the approach. Targeted rail and pipeline investments reduce total transportation costs and cut CO₂ emissions by ~ 27.5% compared to the cost-only optimum. Out-of-sample tests across different uncertainty scenarios confirm the robust model's superiority. It achieves 100% feasibility, vs. 50% for the nominal model, lowers average cost by 4.7%, reduces average emissions by 30.4%, and improves uncertainty regret indices by up to ~ 92%. These findings highlight the model's resilience and ability to deliver sustainable, cost-effective petroleum logistics under real-world uncertainty.
Vibration-damping tracks of an urban rail transit vibrate under train loads and are a source of acoustic sound radiation in urban rail transit. In addition, they exhibit their own vibration amplification phenomenon during service. To evaluate the acoustic properties of damping tracks, this study considered acoustic wave superposition (AWSM) and examined a steel-spring floating-slab track (SSFST). This study focused on floating-slab (FS) acoustic radiation prediction and proposed a fast prediction method for the acoustic radiation of a track structure with a more regular shape. The influence of the structural parameters of FS on its acoustic characteristics was analysed and suggestions regarding the application of acoustic vibration characteristics were provided. The steady-state frequency-domain acoustic radiation prediction method exhibited high solution efficiency compared with the acoustic boundary element method, and was combined with the vehicle-track coupled dynamics theory and AWSM. The proposed method can be used to predict the acoustic radiation characteristics of regular structures and quickly evaluate the acoustic radiation effects. In addition, the FS has a strong low-frequency acoustical sound radiation ability, rendering it an important low-frequency acoustical sound radiation source for urban rail transit. The length and thickness of the FS significantly affected its acoustic vibration characteristics; therefore, a longer FS can be considered for sections with higher requirements for low-frequency noise. In the standard range, selecting a thicker FS offers advantages in terms of the acoustic sound radiation and vibration characteristics.
Arc sprayed Zn15Al coatings with different spraying parameters: arc current and spraying distance were investigated. The corrosion behavior of the coatings was studied using an electrochemical impedance spectroscopy (EIS), cyclic corrosion testing (CCT-4), and coastal railway exposure test. Corroded surfaces were characterized by scanning electron microscopy (SEM). The corrosion product formed on the coating was analyzed by X-ray diffraction (XRD) and scanning electron microscopy. From the potentiodynamic polarization test, the corrosion rate of the coating, using arc current of 100 A and a spraying distance of 100 mm, was 0.012 mm/year. The corrosion rate increased with increasing arc current and spraying distance. The corrosion mechanism of Zn15Al coating started with selective dissolution of Zn-rich phase leaving behind the passivated Al-rich phase and followed by stable corrosion product formation. From the surfaces analysis after the corrosion test, Simonkolleite (Zn5(OH)8Cl2•H2O) was found on the coating surfaces. Similarly, the samples exposed on the coastal railway line revealed Zn-rich phase dissolution and Al-rich phase passivation, but without protective corrosion product formations. Based on the corrosion rates, the results from the coastal railway exposure agreed well with electrochemical impedance measurement and cyclic corrosion test. Suggested spraying parameters for onsite repair was 100 A arc current and 100 mm spraying distance to achieve the optimal corrosion resistance of Zn15Al coating.
Carbon nanotubes (CNTs), combining excellent electrical and optoelectronic properties with low-temperature processability, provide a compelling materials platform for monolithic three-dimensional (M3D) integration that unifies digital logic in complementary field-effect transistor (CFET) architecture and functional sensing elements with three-dimensionally structured nondigital functional blocks. However, such a fully integrated system has not yet been experimentally demonstrated. Here, we report CNT-based digital circuits implemented in a true CFET architecture, in which vertically stacked P- and N-FETs share an identical footprint and exhibit well-balanced performance through structural engineering. A full suite of logic functions, including inverters, NOR, OR, NAND, AND gates, as well as a 4-transistor static random-access memory cell and a five-stage ring oscillator are successfully demonstrated. The CFET inverters exhibit rail-to-rail operation with large noise margins and a peak voltage gain of 147 at a supply voltage of 1 V, while maintaining a gain of 9.7 with only 3.3 pW static power consumption at 0.2 V. In parallel, CNT photodiodes are vertically stacked and cascaded to form a 3D optical sensor that delivers nearly twice the photovoltage of planar counterparts. By monolithically integrating the 3D CNT photodiode with a CNT-based CFET inverter, we further demonstrated a prototype "sensing-and-computing" module in which optical power and spectral information are directly sensed and processed within a single monolithic CNT-based block. This work establishes CNTs as a unified platform for high-density, low-power M3D integration toward near-/in-sensor and edge-computing applications.
Paravalvular leak (PVL) is a complication after prosthetic valve implantation, leading to issues like hemolysis and heart failure. While redo surgery is the standard treatment, it poses significant risks for high-risk patients. Transcatheter PVL closure has become a less invasive option, with improved efficacy from advanced imaging and device technologies. This study aims to evaluate the safety, efficacy, and technical strategies of transcatheter PVL closure, including novel approaches like the reverse loop and arteriovenous (AV) rail techniques in a prospective single-center cohort. Twenty-eight high-risk patients (mean age 57.2 ± 10.1 years) with significant aortic or mitral PVLs underwent transcatheter closure. Procedures were guided by multimodal imaging, assessing technical, procedural, and clinical success rates, along with 6-month outcomes in NYHA functional class and echocardiographic parameters. Out of 28 procedures, 27 were technically successful (96.4%), with clinical success achieved in 92.9% of patients. Most procedures used transfemoral access, treating aortic PVLs retrogradely and mitral PVLs through various routes. Advanced techniques like reverse loop (n = 4) and AV rail (n = 3) were utilized in complex cases. NYHA functional class improved significantly (p < 0.001), with no significant change in left ventricular ejection fraction. One patient required redo surgery, and one died from sepsis. Transcatheter PVL closure is a safe and effective alternative to redo surgery for selected high-risk patients. Advanced imaging and access techniques enhance procedural success, particularly in complex mitral PVLs. Further multicenter studies with long-term follow-up are needed to confirm these results and refine technique selection.
Segmental trunk control is central to upright motor development in preterm infants, yet evidence linking specific home environment attributes to condition-specific control in late infancy remains limited. This study aims to examine the associations between characteristics of commercial baby containers-defined as restricting equipment or furniture such as playpens and cots-and segmental trunk control during the standing-walking window. Additionally, age-related usage patterns of these containers from 8 to 13 months corrected age were described. In a prospective longitudinal study of 76 moderate-to-late preterm infants, caregivers provided monthly reports on container usage, specifically categorizing attributes of size, rail height, and floor surface compliance. Segmental trunk control was assessed monthly using the Segmental Assessment of Trunk Control (SATCo) under static, active, and reactive conditions. Results indicated that trunk control improved with age across all conditions, although reactive control initially developed more slowly than static and active control before converging at full proficiency by 13 months. Larger container size was consistently linked to better trunk control across static, active, and reactive conditions. Floor surface compliance also played a significant role; soft floors were associated with better static and active control but did not influence reactive control. Conversely, the height of the container rails showed no independent relationship with trunk control performance. Usage patterns shifted over time toward larger spaces and firmer surfaces. These findings suggest that in late infancy, expanded physical space consistently supports segmental trunk control, while surface compliance specifically benefits steady-state and anticipatory control. The results support low-cost, home-based guidance that encourages expanding safe play areas and selecting appropriately compliant surfaces to foster motor development in preterm infants.