Search Results (8,810)

The operation of medical cyclotrons for PET radiopharmaceutical production presents significant radiological and environmental challenges that require systematic risk assessment and evidence-based mitigation strategies. In this study, an integrated framework combining Failure Mode and Effects Analysis (FMEA) with a quantitative Defense Effectiveness Factor (DEF) approach to evaluate and reduce residual risk in a real urban cyclotron facility. High-criticality failure modes (Risk Priority Number $\ge 120$) affecting HVAC systems, stack exhaust, and power supply were identified and validated through a Delphi expert consensus process. These modes were addressed with multi-layered defense-in-depth strategies: redundant systems (occurrence reduction, 60–80% effectiveness), real-time monitoring (detection reduction, 40–50% effectiveness), and design robustness (severity reduction, 70–85% effectiveness). The combined DEF yielded a 96–97% risk reduction. One-way sensitivity analysis confirmed the robustness of these results, with residual annual effective dose to the representative person remaining between 50–88 $\mathsf{\mu }$Sv/year (well below the IAEA 1 mSv/year public dose constraint) even under pessimistic scenarios. Primary exposure pathways were inhalation and cloud gamma from ${}^{18}$F and ${}^{41}$Ar during the early-morning production window, while secondary pathways were negligible due to the short half-lives of the radionuclides. These findings demonstrate that the integration of FMEA with DEF-based defense-in-depth and Gaussian plume modeling provides a transparent, robust, and regulatory-compliant framework for managing radioactive atmospheric emissions in urban medical cyclotron facilities. Full article

Infrared small target detection (ISTD) plays a crucial role in many real-world applications. However, this task remains highly challenging due to the extremely small target size, low contrast, and complex background interference as infrared small targets often occupy fewer than 80 pixels in a $256\times 256$ image under a commonly used ISTD criterion. Although Segment Anything Model (SAM) shows strong generalization in image segmentation, directly applying SAM to ISTD is suboptimal, primarily due to the significant modality gap between RGB and infrared imagery, as well as the prohibitive cost of full-parameter fine-tuning. To address these challenges, we propose a prompt-free and parameter-efficient fine-tuning framework that adapts SAM for ISTD. To bridge the cross-modality gap while preserving the pretrained prior knowledge of SAM, a lightweight Infrared Adapter (IR-Adapter) is introduced into the image encoder, enabling effective task adaptation with only a small number of trainable parameters. Furthermore, to alleviate the loss of small target information in deep network layers, we design a Multi-Scale Feature Fusion (MSF) module that integrates hierarchical features from different encoder stages. In addition, a Coarse-to-Fine Head (CFH) with dual-branch prediction is proposed to incorporate fine-grained details for more accurate target localization and segmentation. Extensive experiments conducted on two public datasets demonstrate that the proposed method achieves better overall performance than existing representative approaches, yielding higher IoU, nIoU and Pd. Full article

Ormosia macrocalyx grows in tropical forests and is endangered in Mexico. The species has ecological and economic importance. To evaluate the relationship between fruit length and seed number, Pearson correlation and principal component analysis were used. A linear mixed-effects model was also applied. Pearson correlation, principal components analysis (PCA) and a Linear Mixed-Effects Model (LMM) were performed on an exploratory basis. In addition, seed coat and cotyledon anatomy were examined, and histochemical tests for secondary metabolites were carried out. Two high correlations and two components were obtained from the PCA, and the LMM showed that fruit length influenced the number of seeds per fruit. In the seed coats, differentiated layers of macrosclereids and osteosclereids were identified, where the hilar region presented macrosclereids and a pyriform bar of tracheids, the reserved cotyledons showed double-walled cells and simple plasmodesmata, the histochemical analyses demonstrated the presence of cellulose, condensed tannins, lipids, alkaloids, and proteins, and no starch was present. This study provides the first description of seed coat and cotyledon anatomy in O. macrocalyx, as well as the first report of secondary metabolites in storage cotyledons. These results could be useful for further studies of this species. Full article

The rapid and accurate assessment of residual ultimate strength after ship damage is crucial for rescue decision-making and navigation safety, while traditional methods struggle to meet the demands of complex random damage scenarios in terms of efficiency or accuracy. This study proposes a hybrid framework that integrates high-fidelity nonlinear finite element simulation (NFEM) and a Bayesian-regularized backpropagation neural network (BPNN). NFEM is used to accurately simulate a large number of random damage scenarios, generating a physically credible benchmark dataset. BPNN serves as an efficient surrogate prediction model, with its key parameters—the number of hidden layers and the training algorithm—systematically optimized to enhance generalization capability. The results show that: (1) The NFEM simulation results deviate by less than 5% compared to the Smith method, validating the reliability of the dataset. (2) The prediction performance of BPNN is highly dependent on the number of hidden layers and the training algorithm, exhibiting non-monotonic variation, with an optimal parameter combination identified as 8 hidden layers paired with the Bayesian algorithm, achieving a prediction regression value R of 0.91662. (3) Deep networks are prone to overfitting, while shallow networks suffer from insufficient feature capture. (4) The Bayesian algorithm performs best in terms of overfitting resistance and stability. This study not only provides a high-precision and efficient intelligent solution for residual strength assessment of damaged hulls, but its systematic neural network parameter optimization strategy, particularly the approach of identifying optimal depth and selecting anti-overfitting algorithms, also offers an important reference for the design of intelligent damage assessment models for similar engineering structures. Full article

Aiming at the problems of the large prediction error of model-driven algorithms and poor interpretability (even potential violation of physical laws) of pure data-driven algorithms in the prediction of aerospace vehicle plume characteristics, a physics mechanism-guided prediction algorithm for aerospace vehicle plume characteristics was proposed. Taking the long short-term memory (LSTM) network as the backbone, this algorithm constructed a hybrid physics–data model by embedding the prior knowledge of physical laws and empirical rules into the neural network, and designed a loss function combined with physical mechanisms to guide network training. The aerospace vehicle plume dataset was preprocessed through characteristic parameter extraction, extended physical parameter calculation, data splicing and sliding window operation, and the LSTM network structure was optimized by adjusting hyperparameters such as the number of hidden layers and neurons. Experimental results show that the proposed algorithm achieves a Mean Absolute Error (MAE) of 31.89 and a Physical Inconsistency of 0.1723 on the test set, with MAE reduced by 14% and Physical Inconsistency reduced by 7.5% compared with traditional machine learning models such as Random Forest. Ablation experiments verify that the introduction of physical mechanisms can improve the prediction accuracy of the model by about 25%. This algorithm makes up for the defects of traditional prediction algorithms, has good generalization ability and physical consistency, and provides an effective method for the prediction of engine exhaust plume temperature distribution. Full article

Limited ionic conductivity and unstable interfaces, primarily caused by poor solid–solid contact, pose significant challenges to the stable cycling of solid-state batteries. In this study, an interfacial in situ polymerization strategy is proposed to construct a poly(1,3-dioxolane) (PDOL) gel electrolyte layer between a poly(vinylidene fluoride) (PVDF)-based solid polymer electrolyte and the electrodes. This approach aims to address interfacial compatibility issues in solid-state lithium metal batteries. By precisely tuning the composition of the gel precursor and employing characterization techniques such as FTIR and NMR, the efficient ring-opening polymerization of 1,3-dioxolane (DOL) was confirmed, achieving a high conversion rate of 90%. The precursor was drop-cast onto the PVDF-based electrolyte/electrode interfaces before cell assembly. Electrochemical evaluations revealed that the in situ formed solidified interlayer significantly enhanced interfacial compatibility and ion transport, yielding a high Li⁺ transference number (0.341), an exceptional critical current density (1.4 mA cm⁻²), and remarkable cycling stability exceeding 1600 h in Li||Li symmetric cells. Furthermore, full cells incorporating LiFePO₄ cathodes demonstrated excellent rate capability and long-term cyclability, retaining 98.7% of their capacity after 1000 cycles. These results collectively underscore the effectiveness of this in situ solidification strategy in optimizing the interface structure and improving the overall performance of PVDF-based solid-state batteries. Full article

Bingham fluids, also called Bingham plastics, are used in different industries including the production of food, pharmaceuticals, household products, construction and oil and gas drilling. The behavior of Bingham fluids is viscous above a critical shear stress and rigid-body below the threshold stress value. Knowledge of the size of the entrance region has several applications including hemodynamics and microfluidics. A model for steady Bingham fluid flow in the entrance region between parallel plates is developed using the inlet-filled region concept. A boundary layer model is used to solve the fluid flow dynamics in the inlet region up to the point where the critical shear stress is reached at the edge of the boundary layer. Beyond that point, the boundary layer does not grow, while the velocity profile keeps readjusting in the filled region to asymptotically reach the fully developed flow. The results include boundary layer thickness profiles, dimensionless pressure drop, centerline velocity, friction factor and inlet and entrance region sizes as functions of the Bingham number. The results are validated against the results for the Newtonian fluid case (Bingham fluid yield stress equal to zero) and CFD results, using the finite element method, for nonzero Bingham numbers. In addition, the results are found to asymptotically reach the fully developed flow values for the general Bingham fluid flow case. The effects of the Bingham number are addressed and compared with the literature. The present model is largely analytical, requiring minor numerical tasks. Full article

Ochetobibus elongatus (Kner) is a migratory fish found in the Yangtze River basin and areas south of it, and listed as a critically endangered (CR) fish on the China Red List of Vertebrates. To achieve group recovery and artificial breeding, this study investigated the dietary characteristics of O. elongatus based on high-throughput sequencing of its intestinal contents, and its digestive system morphology, and its histology. Results showed that the digestive system of O. elongatus lacked a stomach and mainly consisted of the oropharynx, pharyngeal teeth, esophagus, intestine, and anus. The gut index was 0.88, with clear segmentation of the foregut, midgut, and hindgut, and the visceral mass index was 7.35%. Histological analysis of the digestive system revealed the presence of keratinized dental plates or pharyngeal teeth in the pharynx, as well as a high density of taste bud cells in the soft palate of the oral cavity. The surface layer of the intestinal villi contained numerous mucous cells, with the average number of mucous cells per villus gradually increasing from the esophagus to the hindgut, and the foregut having the longest and most abundant mucosal folds. The esophagus exhibited well-developed circular and longitudinal muscle layers, while in the hindgut, both the circular and longitudinal muscle layers were slightly thicker than those in the midgut. High-throughput sequencing of the intestinal contents of O. elongatus revealed the following phyla based on 18S V4 meta-barcoding: Chlorophyta, Diatoms, Arthropoda, Basidiomycetes, and Ascomycetes, with the genus Hypophthalmichthys and algae being the main classifications. In contrast, based on COI meta-barcoding, the study newly identified the phyla Cnidaria and Mollusca, with the genera Chlorophyta, Scenedesmus, Pectinodesmus, and zooplankton such as Pseudodiaptomus. Metagenomic sequencing revealed that the gut microbiota at the phylum level was predominantly composed of Pseudomonadota, Ascomycota, Basidiomycota, Chytridiomycota, and Bacillota, with key genera including Cetobacter, Pseudomonas, Acinetobacter, Aeromonas, and Clostridium. This study indicates that O. elongatus is an omnivore with carnivorous tendencies. Basic biological research on O. elongatus is of great significance for the restoration of the population, artificial breeding, and the development of its artificially formulated feed. It also provides important data for the formulation of biodiversity conservation measures. Full article

Inherently, there exists a significant security hole in sensor networks. The majority of sensors are not high-end Internet of Things (IoT) devices with sufficient computing resources. Connected sensors (physical nodes in real networks) are allocated to logical nodes and managed remotely by a supervisor in a virtual network. Data acquired by sensors are then collected by a data center on which artificial intelligence operates. If an adversary spoofs a logical node (e.g., an account in a transport layer security (TLS) session) of a vulnerable sensor on the network, then it can manipulate data input to artificial intelligence. Artificial intelligence cannot verify the integrity of the data input for learning. It is difficult to stop data poisoning with no countermeasures against session spoofing. To avoid session spoofing, physical and logical nodes must be linked seamlessly. One might think this can be achieved by utilizing Hardware Root-of-Trust (HRoT) based on a Physically Unclonable Function (PUF). However, a PUF is based on an expensive System-on-a-Chip (SoC), which has been specifically designed for high-end devices, like expensive smartphones. Many sensors (low-end and middle-end IoT devices) can hardly be protected with existing PUFs. Since the number of IoT devices with a PUF is insufficient to cover the entirety of IoT devices, an attacker can find a vulnerable IoT device with no PUF to perform session spoofing. This is the problem of numbers. To resolve it, we propose Physical Cyber Authentication (PCA). A Blockchain account (a logical node in a TLS session) is anchored to an integrated circuit (IC) chip inside a sensor, allowing Blockchain to manage sensor networks, which provides necessary data to artificial intelligence, thus forming a Blockchain of sensors. Full article

Background/Objective: Smoking is known to be associated with reflux-related mucosal damage and deleterious esophageal outcomes, yet no non-invasive imaging biomarkers of smoking-induced esophageal remodeling have been identified. We aimed to compare cervical esophageal ultrasound morphology between habitual smokers and non-smokers, in terms of esophageal wall thickness, number of sonographically discernable wall layers, and esophageal diameter, and investigate whether smoking is an independent predictor of these findings. Methods: In this cross-sectional study, 60 participants (30 smokers, 30 non-smokers) underwent high-resolution B-mode ultrasound of the cervical esophagus. Examinations were performed in transverse and longitudinal planes. Outcomes included esophageal wall thickness (mm), number of discernible wall layers, and esophageal diameters in transverse and longitudinal planes. Group comparisons used independent t-tests and chi-square tests. Multiple linear regression assessed independent associations with smoking status (adjusting for age and weight). Within smokers, Pearson correlation evaluated relationships between smoking duration and ultrasound outcomes; exploratory subgroup analyses compared smoking modalities. Results: Smokers were older and had higher weight and BMI than non-smokers. Compared with non-smokers, smokers had greater wall thickness (3.06 vs 2.61 mm), more discernible wall layers (5.03 vs 3.60), and larger transverse (11.68 vs 7.87 mm) and longitudinal (12.90 vs 8.26 mm) diameters (all p < 0.001). In regression analysis, smoking status independently predicted wall thickness (B = 0.411 mm, 95% CI 0.243–0.578; p < 0.001). Smoking duration showed significant correlations with the number of visible layers (r = 0.82; p < 0.001) and wall thickness (r = 0.42; p = 0.021). Conclusions: High-frequency ultrasound detected significant differences in cervical esophageal morphology between smokers and non-smokers. Smoking was independently associated with differences in the diameter, thickness, and number of visible layers of the cervical esophagus. Further studies with larger sample sizes, improved exposure assessment, and use of reference standards are needed. Full article

As an important state parameter in battery management systems, accurate state of charge (SOC) estimation is of great significance for the safe and reliable use of batteries. In this paper, a Temporal Convolutional Network–Transformer (TCN–Transformer) model is proposed for achieving accurate estimation of SOC. First, the TCN is integrated in series with the Transformer model. This integration not only extracts the local characteristics of time-series data but also captures broader spatiotemporal correlations, thereby enhancing the feature representation and achieving highly accurate estimation. However, since the hyperparameter settings of neural networks have a significant impact on model performance, this study employs the advanced hippo optimization (HO) algorithm to determine the optimal values for the number of filters, filter size, number of residual blocks, and number of encoder layers, ultimately improving the model’s stability and efficiency. Finally, the proposed model was tested under various dynamic driving conditions at different temperatures. Experimental validation on the CALCE dataset demonstrates that the proposed HO–TCN–Transformer achieves RMSE and MAE both under 0.7%, representing an approximately 50% overall error reduction compared to the standalone TCN. Cross-validation across five folds confirms robust performance with <7% standard deviation. Full article

To improve the assembly efficiency and productivity of complex aircraft components, the optimization of an assembly line was investigated in this study. A hierarchical hybrid multi-objective optimization algorithm (HHMOA) was proposed using an improved non-dominated sorting genetic algorithm II and an enhanced longest processing time algorithm. The algorithm incorporates a two-layer framework for global–local optimization; an information entropy-based problem formulation with three objectives, including line balance rate, load balance index and assembly complexity smoothness index; and a hybrid initialization strategy for high-quality initial solutions. Based on the assembly line datasets of different scales, the algorithm performance was verified by comparing the hypervolume and the calculation efficiency using HHMOA and three benchmark algorithms, and the sensitivity analyses verified the algorithm robustness. For an actual aircraft component assembly line, the optimizations carried out with the given process time, number of workstations and precedence relationships indicate that the balance rate of the optimized line increased 72%, and the load balance index and the assembly complexity smoothing index were reduced by 80.3% and 92% respectively, which proved the reliability of the hybrid algorithm in optimizing the aircraft component assembly line. Finally, the optimization analyses with various workstation numbers and assembly process times suggest that reducing the workstations and adopting robotic automated processing can improve the aircraft component assembly line. Full article

Purpose: To explore anatomically adjusted point-to-point relationships between macular sensitivity and intraretinal layer thickness in healthy and glaucomatous eyes using combined optical coherence tomography (OCT) and microperimetry. Methods: Seventy-two eyes were included (27 healthy controls and 45 eyes with primary open-angle glaucoma). Retinal sensitivity was assessed using MP-1 microperimetry, and retinal structure was evaluated with Spectralis OCT. Automatic segmentations included macular retinal nerve fiber layer (mRNFL), ganglion cell layer (GCL), inner plexiform layer (IPL), GCL + IPL, ganglion cell complex (mRNFL + GCL + IPL), and the outer retinal layer. Microperimetry maps were anatomically aligned with OCT grids using vascular landmarks, and ganglion cell displacement was considered when analyzing inner retinal layers. Thickness measurements were obtained at corresponding anatomical points, and structure–function associations were assessed using Spearman correlation analysis to generate spatial correlation maps. Results: Almost no significant pointwise correlations were detected in healthy eyes across any retinal segmentation. In glaucomatous eyes, significant positive correlations were observed for inner retinal layers, whereas no significant associations were found for the outer retinal layer. Distinct spatial patterns were identified, with peripheral correlations for mRNFL and paracentral temporal correlations for GCL, IPL, and GCL + IPL. The highest number of significant associations was observed for the ganglion cell complex. Conclusions: Anatomically adjusted pointwise analysis revealed localized and heterogeneous patterns of macular structure–function coupling predominantly involving ganglion cell-related layers in glaucoma. High-resolution mapping may uncover spatial relationships that are partially obscured by regional or spatially averaged approaches and should be interpreted as a complementary exploratory strategy rather than a replacement for established regional analyses. Full article

Edge computing has emerged as a promising paradigm to minimize latency and energy consumption while improving computational efficiency for mobile devices. Latency-sensitive applications such as autonomous driving, augmented reality, and industrial automation require ultra-low response times, making efficient task offloading a necessity in edge computing. However, distributing optimally computational tasks among edge servers remains a challenge, especially when considering latency, energy consumption, and workload balancing simultaneously. Although existing approaches have focused on one or two of these objectives, they do not provide a holistic solution that incorporates all three factors. In addition, some existing solutions do not take advantage of parallelism at the edge layer, resulting in bottlenecks and inefficient resource usage. In this paper, we propose a novel learning-based task offloading model that integrates parallel processing at the edge layer, adaptive workload balancing, and joint latency–energy optimization. Moreover, by dynamically adjusting the number of selected edge servers for parallel execution, our approach achieves optimal trade-offs between performance and resource efficiency. Our experimental setup includes several edge servers and several randomly deployed devices. It employs Apache HTTP Benchmark (AB) to generate realistic Mobile Edge Computing workloads. The obtained results show that our method outperforms existing approaches by reducing latency, lowering energy consumption, and maintaining a balanced workload across edge nodes. Full article

The need to create lightweight materials with better mechanical properties has led to the use of Fiber Reinforced Composites (FRCs)s in the aerospace and automotive industries. The mechanical behavior of FRCs is heterogeneous, especially in conditions of low-velocity impact (LVI). The impact events cause structural damage, where most of the available literature deals with mono- or bi-composites in controlled situations. This work will present the results of studying the behavior of mono, bi- and tri-hybrids with carbon, glass and Kevlar fiber-reinforced epoxy. The sequences of the laminate stacks, number of plies and laminate thickness in the drop weight testing were across velocities of 1.91 to 3.91 m/s at drop heights of 19 to 79 cm. The dominant pillars of LVI, such as peak load, energy absorption and the modes of damage, were analyzed. The glass-dominated laminates peaked at 5.67 kN, while the Kevlar-dominated laminates reached peak flow in ductile collapse with greater quantities of absorbed energy. The leaders in strength and energy were the hybrids of Kevlar–glass (KG) cross-ply at 8.08 kN and 47.28 J and quasi-isotropic Kevlar–carbon–glass (KCG) at 9.12 kN and 47.25 J, showcasing a balance of strength and toughness. The rest, holding a greater quantity of Kevlar, ranging in thickness and cross-plies, were shaped with a load center. The experimental conclusion is that hybridization improved impact resistance and ductility, which is best supported by the glass/carbon rigidity-layered laminates. Such understanding directs the design work of future composite materials for better impact control. Full article