Search

• Reliability-based non-probabilistic multiscale topology optimization (NRBMTO) with stress constrained is proposed. • The p-norm function aggregates global unit stresses. • Mechanical and thermo-mechanical loads are considered as non-probabilistic uncertain parameters. • Ellipsoid modeling describes uncertainty in non-probabilistic random variables. • NRBMTO results in more security than the traditional deterministic multiscale topology optimization (DMTO) approach. A reliability-based non-probabilistic multiscale topology optimization (NRBMTO) method with stress constraints is proposed for thermo-mechanical continuous structures with uncontrollable stresses. The physical parameters, external loads and temperature values at the macro scale, are regarded as non-probabilistic uncertain parameters in the optimization of structural topologies with complex physical fields at multi-scale. The homogenization-based finite element method is employed to quantify thermo-mechanical structures with multi-scale uncertain parameters in the established multi-scale topology model. The ellipsoid model is applied to describe the uncertainty of non-probabilistic random variables, and the non-probabilistic reliability index is obtained by estimating the failure probability based on the first-order reliability method (FORM). The unit stresses are aggregated to the global maximum stresses with the normalized p-norm function, taking into account the mechanical and thermal stresses. The sensitivity information of the compliance and stress constraint to the macro- and micro-design variables and uncertain variables are derived simultaneously. The macro- and micro- design variables are solved by the method of moving asymptotes (MMA), respectively. Several numerical examples are given to verify the effectiveness and feasibility of the proposed NRBMTO method. The results demonstrate that the optimized structure based on the NRBMTO method provides better security with reliability index β =3 and minimum compliance (244.39) while stress is controlled below 235 MPa compared to the classical deterministic multiscale topology optimization (DMTO) method. [ABSTRACT FROM AUTHOR]

Copyright of Journal of South-Central Minzu University (Natural Science Edition) is the property of Journal of South-Central Minzu University (Natural Science Edition) Editorial Office and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

In recent years, with the continuous development of autonomous driving, monocular 3D object detection has garnered increasing attention as a crucial research topic. However, the precision of 3D object detection is impeded by the limitations of monocular camera sensors, which struggle to capture accurate depth information. To address this challenge, a novel Aggregation Transformer Network (ATNet) is introduced, featuring Cross-Attention based Positional Aggregation and Dual Expansion-Squeeze based Channel Aggregation. The proposed ATNet adaptively fuses radar and camera data at both positional and channel levels. Specifically, the Cross-Attention based Positional Aggregation leverages camera-radar information to compute a non-linear attention coefficient, which reinforces salient features and suppresses irrelevant ones. The Dual Expansion-Squeeze based Channel Aggregation utilizes refined processing techniques to integrate radar and camera data adaptively at the channel level. Furthermore, to enhance feature-level fusion, we propose a multi-scale radar-camera fusion strategy that integrates radar information across multiple stages of the camera subnet's backbone, allowing for improved object detection across various scales. Extensive experiments conducted on the widely-used nuScenes dataset validate that our proposed Aggregation Transformer, when integrated into superb monocular 3D object detection models, delivers promising results compared to existing methods. [ABSTRACT FROM AUTHOR]

In the field of image object detection and semantic segmentation, improving the accuracy of object identification and segmentation is a primary goal. To achieve this, leveraging the potential of multi-scale information through feature map refinement and fusion has been widely recognized. However, existing feature fusion methods either design more complex feature pyramid networks, replace existing detectors, or incrementally introduce feature fusion modules, overlooking the effective approach of enhancing spatial information in deep feature maps. We propose a novel pluggable feature fusion paradigm termed ‘Effective Learning Bridge’. Our research introduces an efficient and adaptive learning mechanism that builds learning bridges between feature maps at different scales within the feature pyramid, thereby enhancing the spatial information of objects in deep feature maps. This mechanism is specifically designed for multi-scale feature maps and can be seamlessly integrated into any network incorporating feature maps. By altering the model’s backpropagation path, we successfully improve learning efficiency, which in turn enhances the accuracy of object detection and segmentation. Our proposed paradigm and method were extensively evaluated through experiments on SIMD, HRSID, and WHDLD datasets and benchmark models. The results unequivocally demonstrate the effectiveness of our approach in significantly improving the accuracy of object detection and semantic segmentation, as well as the overall learning efficiency of the model. [ABSTRACT FROM AUTHOR]

Organic photovoltaics (OPVs) have made significant strides with efficiencies now exceeding 20%, positioning them as potential competitors to inorganic solar technologies. One of the most critical challenges toward this goal is the severe open‐circuit voltage (Voc) loss caused by the nonradiative charge recombination (NRCR). Herein, this review comprehensively summarizes the NRCR mechanisms and suppression techniques of OPVs across various scales from molecule to device. Specifically, the origins of NRCR in a single molecule are first summarized, and molecular design principles toward high photoluminescence quantum yield are reviewed following the Marcus theory. Next, the effect of aggregation on NRCR is reviewed, as well as the molecular and processing strategies to modulate the film packing for NRCR suppression. Furthermore, the progresses in the avoidance of nonradiative loss pathways mediated by charge transfer states and triplet states in donor:acceptor bulk heterojunctions are tracked. Besides, the interfacial optimization and device structure design to maximize the electroluminescent quantum efficiency are presented. Finally, several potential pathways toward curtailing NRCR for high‐performance OPVs are outlined. Therefore, this review shows an insightful perspective to understand and mitigate the NRCR at multi‐scales, and is poised to provide a clear roadmap for the next breakthrough of OPVs. [ABSTRACT FROM AUTHOR]

Raindrops can scatter and absorb light, causing images to become blurry or distorted. To improve image quality by reducing the impact of raindrops, this paper proposes a novel generative adversarial network for image de-raining. The network comprises two parts: a generative network and an adversarial network. The generative network performs image de-raining. The adversarial network determines whether the input image is rain-free or de-rained. The generative network comprises two branches: the A branch, which follows a traditional convolutional network structure, and the U branch, which utilizes a U-Net architecture. The A branch includes a multi-scale module for extracting information at different scales and a residual attention module to reduce redundant information interference. The U branch contains an encoder module designed to address the loss of details and local information caused by conventional down-sampling. To improve the performance of the generative network in image de-raining, this paper employs a relative discriminator incorporating a mean squared error loss. This discriminator better measures the differences between rainy and rain-free images while effectively preventing the occurrence of gradient vanishing. Finally, this study performs visual and quantitative comparisons of the proposed method and existing methods on three established rain image datasets. In the quantitative experiments, the proposed method outperforms existing methods regarding PSNR, SSIM, and VIF metrics. Specifically, our method achieves an average PSNR, SSIM, and VIF of approximately 5%, 3%, and 4% higher than the MFAA-GAN method, respectively. These results indicate that the de-rained images generated via the proposed method are closer to rain-free images. [ABSTRACT FROM AUTHOR]

System‐scale computational fluid dynamics (CFD) simulations in chemical and process engineering remain limited owing to the complexity of integrating the results obtained at different scales. The present study addresses this issue by correlating the flow behaviors calculated by CFD in porous media at the micro‐scale and the macro‐scale using deep neural network (DNN) technology. The DNN model is trained using a dataset constructed from the results obtained for a large number of particle‐scale CFD simulations that are coupled to macroscopic governing equations. Comparisons with experimental results obtained with a packed bed show that the proposed CFD‐DNN method provides predictions of pressure drop with an accuracy that is 28% greater than that of a method based on the Ergun equation. [ABSTRACT FROM AUTHOR]

This work presents the PULPO (Python‐based user‐defined lifecycle product optimization) framework, developed to efficiently integrate life cycle inventory (LCI) models into life cycle product optimization. Life cycle optimization (LCO), which has found interest in both the process systems engineering and life cycle assessment (LCA) communities, leverages LCA data to go beyond simple assessments of a limited number of alternatives and identify the best possible product systems configuration subject to a manifold of choices, constraints, and objectives. However, typically, aggregated inventories are used to build the optimization problems. Contrary to existing frameworks, PULPO integrates whole LCI databases and user inventories as a backbone for the optimization problem, considering economy‐wide feedback loops between fore‐ and background systems that would otherwise be omitted. The open‐source implementation combines functions from Brightway2 for the manipulation of inventory data and pyomo for the formulation and solution of the optimization problem. The advantages of this approach are demonstrated in a case study focusing on the design of optimal future global green methanol production systems from captured CO2 and electrolytic H2. It is shown that the approach can be used to assess sector‐coupling with multi‐functional processes and prospective background databases that would otherwise be impractical to approach from a standalone LCA perspective. The use of PULPO is particularly appealing when evaluating large‐scale decisions that have a strong impact on socioeconomic systems, resulting in changes in the technosphere on which the background system is based and which is often assumed constant in standard LCO approaches regardless of the decisions taken. This article met the requirements for a gold‐gold JIE data openness badge described at http://jie.click/badges. [ABSTRACT FROM AUTHOR]

The Spherical Tokamak for Energy Production (STEP) programme is an ambitious but challenging endeavour to design and deliver a prototype fusion power plant. It is a rapid, fast-moving programme, designing a first of a kind device in a Volatile, Uncertain, Complex and Ambiguous (VUCA) environment, and digital tools play a pivotal role in managing and navigating this space. Digital helps manage the complexity and sheer volume of information. Advanced modelling and simulation techniques provide a platform for designers to explore various scenarios and iteratively refine designs, providing insights into the intricate interplay of requirements, constraints and design factors across physics, technology and engineering domains and aiding informed decision-making amidst uncertainties. It also provides a means of building confidence in the new scientific, technological and engineering solutions, given that a full-scale-integrated precursor test is not feasible, almost by definition. The digital strategy for STEP is built around a vision of a digital twin of the whole plant. This will evolve from the current digital shadow formed by system architecting codes and complex workflows and will be underpinned by developing capabilities in plasma, materials and engineering simulation, data management, advanced control, industrial cybersecurity, regulation, digital technologies and related digital disciplines. These capabilities will help address the key challenges of managing the complexity and quantity of information, improving the reliability and robustness of the current digital shadow and developing an understanding of its validity and performance. This article is part of the theme issue 'Delivering Fusion Energy – The Spherical Tokamak for Energy Production (STEP)'. [ABSTRACT FROM AUTHOR]

In recent years, due to the wide application of multi-sensor vision systems, multimodal image acquisition technology has continued to develop. Monomodal images cannot meet the needs of image analysis, which requires image fusion and stitching to process images for better image analysis. Image registration is an important prerequisite of image fusion and stitching. Most of the existing multimodal image registration methods are only suitable for two modalities, and cannot uniformly register multimodal image data. Therefore, this paper proposes a multimodal remote sensing image registration method based on adaptive multi-scale PIIFD (AM-PIIFD). This method extracts KAZE features in the scale space constructed by nonlinear diffusion filtering. It can effectively preserve the edge feature information while filtering out the noise. Then, the proposed AM-PIIFD feature descriptor is used to describe the multi-scale features. Finally, according to the consistency of the feature main orientation, most of the mismatches are removed, and the image alignment transformation is realized. The qualitative and quantitative comparisons with the other three advanced methods show that our method can achieve good performance in multimodal remote sensing image registration. [ABSTRACT FROM AUTHOR]

In intelligent transportation systems (ITS), achieving accurate travel time estimation (TTE) is paramount, much like route planning. Precisely predicting travel time across different urban areas is vital, and an essential requirement for these privileges is having fine‐grained knowledge of the city. In contrast to prior studies that are restricted to coarse‐grained data, we broaden the scope of traffic flow forecasting to fine granularity, which provokes explicit challenges: (1) the prevalence of inter‐grid transitions within fine‐grained data introduces complexity in capturing spatial dependencies among grid cells on a global scale. (2) stemming from dynamic temporal dependencies. To address these challenges, we propose the multi‐scaling hybrid model (MSHM) as a novel approach. Initially, a multi‐directional convolutional layer is first used to acquire high‐level depictions for each cell to retrieve the semantic attributes of the road network from local and global aspects. Next, we incorporate the characteristics of the road network and coarse‐grained flow features to regularize the local and global spatial distribution modeling of road‐relative traffic flow using an enhanced deep super‐resolution (EDSR) technique. Benefiting from the EDSR method, our approach can generate high‐quality fine‐grained traffic flow maps. Furthermore, to continuously provide accurate TTE over time by leveraging well‐designed multi‐scale feature modeling, we incorporate a multi‐scale feature expression of each road segment, capturing intricate details and important features at different scales to optimize the TTE. We conducted comprehensive trials on two real‐world datasets, BJTaxi and NYCTaxi, aiming to achieve superior results compared to baseline methods. [ABSTRACT FROM AUTHOR]

Automatically extracting roads from remote sensing images is increasingly important for autonomous driving and geographic information systems. However, due to factors such as ground objects blocking road boundaries and some roads being narrow and long in high-resolution remote sensing images, it is difficult for the current mainstream pixel-level extraction model to guarantee the continuity of roads and the smoothness of boundaries. To solve this problem, this paper designs a neural network Rse-net based on semantic segmentation, using a two-stream semantic segmentation network algorithm to make the network pay more attention to the boundary information of roads and narrow roads. This structure uses shape stream to process road boundary information (roads' shape stream), and processes it in parallel with classic stream. Use the Gated Shape CNN to connect the two streams of roads' shape stream and classic stream, and use the classical stream to optimize the boundary information in the shape stream. At the same time, a multi-scale convolutional attention mechanism is used between the decoder and the encoder to expand the receptive field through large-core attention, and obtain more semantic information without causing too much calculation. Finally, the effectiveness of the proposed network is verified by comparative experiments. [ABSTRACT FROM AUTHOR]

People process things and express feelings through actions, action recognition has been able to be widely studied, yet under-explored. Traditional self-supervised skeleton-based action recognition focus on joint point features, ignoring the inherent semantic information of body structures at different scales. To address this problem, we propose a multi-scale Motion Contrastive Learning of Visual Representations (MsMCLR) model. The model utilizes the Multi-scale Motion Attention (MsM Attention) module to divide the skeletal features into three scale levels, extracting cross-frame and cross-node motion features from them. To obtain more motion patterns, a combination of strong data augmentation is used in the proposed model, which motivates the model to utilize more motion features. However, the feature sequences generated by strong data augmentation make it difficult to maintain identity of the original sequence. Hence, we introduce a dual distributional divergence minimization method, proposing a multi-scale motion loss function. It utilizes the embedding distribution of the ordinary augmentation branch to supervise the loss computation of the strong augmentation branch. Finally, the proposed method is evaluated on NTU RGB+D 60, NTU RGB+D 120, and PKU-MMD datasets. The accuracy of our method is 1.4–3.0% higher than the frontier models. [ABSTRACT FROM AUTHOR]

Hybrid networks, benefiting from both CNNs and Transformers architectures, exhibit stronger feature extraction capabilities compared to standalone CNNs or Transformers. However, in hybrid networks, the lack of attention in CNNs or insufficient refinement in attention mechanisms hinder the highlighting of target regions. Additionally, the computational cost of self-attention in Transformers poses a challenge to further improving network performance. To address these issues, we propose a novel point-to-point Dynamic Attention Guider(DAG) that dynamically generates multi-scale large receptive field attention to guide CNN networks to focus on target regions. Building upon DAG, we introduce a new hybrid network called the Dynamic Attention Guider Network (DAGN), which effectively combines Dynamic Attention Guider Block (DAGB) modules with Transformers to alleviate the computational cost of self-attention in processing high-resolution input images. Extensive experiments demonstrate that the proposed network outperforms existing state-of-the-art models across various downstream tasks. Specifically, the network achieves a Top-1 classification accuracy of 88.3% on ImageNet1k. For object detection and instance segmentation on COCO, it respectively surpasses the best FocalNet-T model by 1.6 A P b and 1.5 A P m , while achieving a top performance of 48.2% in semantic segmentation on ADE20K. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Frontiers of Computer Science & Technology is the property of Beijing Journal of Computer Engineering & Applications Journal Co Ltd. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

The mechanical properties of skin change during aging but the relationships between structure and mechanical function remain poorly understood. Previous work has shown that young skin exhibits a substantial decrease in tissue volume, a large macro-scale Poisson's ratio, and an increase in micro-scale collagen fiber alignment during mechanical stretch. In this study, label-free multiphoton microscopy was used to quantify how the microstructure and fiber kinematics of aged mouse skin affect its mechanical function. In an unloaded state, aged skin was found to have less collagen alignment and more non-enzymatic collagen fiber crosslinks. Skin samples were then loaded in uniaxial tension and aged skin exhibited a lower mechanical stiffness compared to young skin. Aged tissue also demonstrated less volume reduction and a lower macro-scale Poisson's ratio at 10% uniaxial strain, but not at 20% strain. The magnitude of 3D fiber realignment in the direction of loading was not different between age groups, and the amount of realignment in young and aged skin was less than expected based on theoretical fiber kinematics affine to the local deformation. These findings provide key insights on how the collagen fiber microstructure changes with age, and how those changes affect the mechanical function of skin, findings which may help guide wound healing or anti-aging treatments. [ABSTRACT FROM AUTHOR]

The Capsule Network (CapsNet) has been shown to have significant advantages in improving the accuracy of bearing fault identification. Nevertheless, the CapsNet faces challenges in identifying the type of bearing fault under nonstationary and noisy conditions. These challenges arise from the distinctive nature of its dynamic routing algorithm and the use of fixed single-scale kernels. To address these challenges, a multi-scale spatial–temporal capsule network (MSCN) based on sequence encoding is proposed for bearing fault identification under nonstationary and noisy environments. A spatial–temporal sequence encoding module focuses on feature correlations at various times and positions. Dilated convolution-based multiscale capsule layer (MCaps) is designed to capture spatial–temporal features at different scales. MCaps establishes connections between various layers, enhancing the comprehension and interpretation of spatial–temporal features. Furthermore, the Bhattacharyya coefficient is introduced into the dynamic routing to compare the similarity between capsules. The validity of the model is verified through comparative experiments, and the results show that MSCN has significant advantages over traditional methods. [ABSTRACT FROM AUTHOR]

This paper addresses the challenge of low detection accuracy of grape clusters caused by scale differences, illumination changes, and occlusion in realistic and complex scenes. We propose a multi-scale feature fusion and augmentation YOLOv7 network to enhance the detection accuracy of grape clusters across variable environments. First, we design a Multi-Scale Feature Extraction Module (MSFEM) to enhance feature extraction for small-scale targets. Second, we propose the Receptive Field Augmentation Module (RFAM), which uses dilated convolution to expand the receptive field and enhance the detection accuracy for objects of various scales. Third, we present the Spatial Pyramid Pooling Cross Stage Partial Concatenation Faster (SPPCSPCF) module to fuse multi-scale features, improving accuracy and speeding up model training. Finally, we integrate the Residual Global Attention Mechanism (ResGAM) into the network to better focus on crucial regions and features. Experimental results show that our proposed method achieves a mAP 0.5 of 93.29% on the GrappoliV2 dataset, an improvement of 5.39% over YOLOv7. Additionally, our method increases Precision, Recall, and F1 score by 2.83%, 3.49%, and 0.07, respectively. Compared to state-of-the-art detection methods, our approach demonstrates superior detection performance and adaptability to various environments for detecting grape clusters. [ABSTRACT FROM AUTHOR]

This paper aims to develop a nonrigid registration method of preoperative and intraoperative thoracoabdominal CT images in computer-assisted interventional surgeries for accurate tumor localization and tissue visualization enhancement. However, fine structure registration of complex thoracoabdominal organs and large deformation registration caused by respiratory motion is challenging. To deal with this problem, we propose a 3D multi-scale attention VoxelMorph (MA-VoxelMorph) registration network. To alleviate the large deformation problem, a multi-scale axial attention mechanism is utilized by using a residual dilated pyramid pooling for multi-scale feature extraction, and position-aware axial attention for long-distance dependencies between pixels capture. To further improve the large deformation and fine structure registration results, a multi-scale context channel attention mechanism is employed utilizing content information via adjacent encoding layers. Our method was evaluated on four public lung datasets (DIR-Lab dataset, Creatis dataset, Learn2Reg dataset, OASIS dataset) and a local dataset. Results proved that the proposed method achieved better registration performance than current state-of-the-art methods, especially in handling the registration of large deformations and fine structures. It also proved to be fast in 3D image registration, using about 1.5 s, and faster than most methods. Qualitative and quantitative assessments proved that the proposed MA-VoxelMorph has the potential to realize precise and fast tumor localization in clinical interventional surgeries. [ABSTRACT FROM AUTHOR]

In this paper, we focus on eddy current array (ECA) technology for defect detection in finely grooved structures of spinning cylinders, which are significantly affected by surface texture interference, lift-off distance, and mechanical dither. Unlike a single eddy current coil, an ECA, which arranges multiple eddy current coils in a specific configuration, offers not only higher accuracy and efficiency for defect detection but also the inherent properties of space and time for signal acquisition. To efficiently detect defects in finely grooved structures, we introduce a spatiotemporal self-attention mechanism to ECA testing, enabling the detection of defects of various sizes. We propose a Multi-scale SpatioTemporal Self-Attention Network for defect detection, called MSTSA-Net. In our framework, Temporal Attention (TA) and Spatial Attention (SA) blocks are incorporated to capture the spatiotemporal features of defects. Depth-wise and point-wise convolutions are utilized to compute channel weights and spatial weights for self-attention, respectively. Multi-scale features of space and time are extracted separately in a pyramid manner and then fused to regress the bounding boxes and confidence levels of defects. Experimental results show that the proposed method significantly outperforms not only traditional image processing methods but also state-of-the-art models, such as YOLOv3-SPP and Faster R-CNN, with fewer parameters and lower FLOPs in terms of Recall and F1 score. [ABSTRACT FROM AUTHOR]

In order to achieve impact load localization of complex structures such as ships, this paper proposes a multi-scale feature fusion convolutional neural network (MSFF-CNN) method for impact load localization. An end-to-end machine learning model is used, where the raw vibration signals of impact loads are directly fed into the network model to avoid the process of feature extraction. Automatic feature learning and feature concatenation of the signal are achieved through four independent convolutional layers, each using a different size of convolutional kernel. Data normalization and L2 regularization techniques are introduced to enhance the data and prevent overfitting. Classification and localization of impact loads are accomplished using a softmax classification layer. Validation experiments are carried out using a ship's stern compartment model. Our results show that the classification and localization accuracy of the impact load sample group of MSFF-CNN reaches 94.29% compared with a traditional CNN. The method further improves the ability of the network to extract state features, takes local perception and global vision into account, effectively improves the classification ability of the model, and has good prospects for engineering applications. [ABSTRACT FROM AUTHOR]

To address the issue of missing near-surface air temperature data caused by the uneven distribution of ground meteorological observation stations, we propose a method for near-surface air temperature estimation based on an improved conditional generative adversarial network (CGAN) framework. Leveraging the all-weather coverage advantage of Fengyun meteorological satellites, Fengyun-4A (FY-4A) satellite remote sensing data are utilized as conditional guiding information for the CGAN, helping to direct and constrain the near-surface air temperature estimation process. In the proposed network model of the method based on the conditional generative adversarial network structure, the generator combining a self-attention mechanism and cascaded residual blocks is designed with U-Net as the backbone, which extracts implicit feature information and suppresses the irrelevant information in the Fengyun satellite data. Furthermore, a discriminator with multi-level and multi-scale spatial feature fusion is constructed to enhance the network's perception of details and the global structure, enabling accurate air temperature estimation. The experimental results demonstrate that, compared with Attention U-Net, Pix2pix, and other deep learning models, the method presents significant improvements of 68.75% and 10.53%, respectively in the root mean square error (RMSE) and Pearson's correlation coefficient (CC). These results indicate the superior performance of the proposed model for near-surface air temperature estimation. [ABSTRACT FROM AUTHOR]

In current convolution-based image dehazing networks, increasing the depth and width of convolutional layers is a common strategy to improve network performance. However, this approach significantly increases the complexity and computational cost of the dehazing network. To address this issue, this paper proposes a U-shaped multi-scale adaptive selection network (UMA-Net). Without introducing additional parameters and computational costs, the network leverages the influence of different scales of convolutional kernels on the receptive field. It combines standard and dilated convolutions into the feed-forward network (FFN) to propose a multi-scale adaptive (MA) dehazing module, further expanding the receptive field and focusing on important spatial and channel information within the FFN. To fully exploit the multi-scale features of the MA module, a lightweight channel attention-guided fusion (CAGF) module is proposed, which achieves the restoration of high-quality dehazed images from hazy images. Extensive experiments demonstrate the effectiveness of the proposed modules. On the Reside SOTS dataset, it achieves state-of-the-art performance with only 0.816M parameters and 8.794G FLOPs. [ABSTRACT FROM AUTHOR]

The study addresses the global impact of COVID-19 by developing a mathematical model that combines within-host and between-host factors to better understand the disease’s dynamics. It begins by describing SARS-CoV-2 dynamics within individual human hosts using fractional-order differential equations. The model is shown to be Ulam–Hyers stable, ensuring reliable predictions. The research then investigates virus transmission from infected to susceptible individuals using agent-based modeling (ABM). This approach allows us to capture the diversity and heterogeneity among individuals, including variations in internal state of individuals, immune response and responses to interventions, making the model more realistic compared to aggregate models. The agent-based model places individuals on a square lattice, assigns health states (susceptible, infectious, or recovered), and relies on infected individuals’ viral load for transmission. Parameter values are stochastically generated via Latin hypercube sampling. The study further explores the impact of viral mutation and control measures. Simulations demonstrate that vaccination substantially reduces transmission but may not eliminate it entirely. The strategy is more effective when vaccinated individuals are evenly distributed across the population, as opposed to concentrated on one side. The research further reveals that while reducing transmission probability decreases infections by implementing prevention protocols, it does not proportionally correlate with the reduction magnitude. This discrepancy is attributed to the intervention primarily addressing inter-host transmission dynamics without directly influencing intra-host viral dynamics. [ABSTRACT FROM AUTHOR]

Global precipitation observations have been realized through the development of satellite remotesensing technology. However, there is a lack of evaluation of remote-sensing precipitation products in complex tropical island terrains. This study used hourly rain gauge data to conduct a multi-scale systematic evaluation of common precipitation products, such as CMORPH, CHIRPS, GsMAP, GPM, MSWEP, ERA5-Land, and PERSIANN, over Hainan Island, providing an in-depth analysis of the precipitation detection capabilities of various products in this region. The main conclusions are: (1) In a multi-temporal scale evaluation, GPM and GsMAP outperformed the other products across all time scales. On a 3-hour scale, GPM and GsMAP showed the highest correlation coefficients (0.53 and 0.52, respectively). On a daily scale, except for PERSIANN, all products showed correlation coefficients above 0.56, with GPM and GsMAP showing the best performance (R = 0.73 and 0.74, respectively). (2) In comparing annual precipitation, Hainan Island's average-annual precipitation over the past 20 years showed a fluctuating trend, with a mean of 1,776.4 mm/a. The CMORPH annual average of 1, 765.1 mm/a was the closest to the CHM-PRE dataset, with minimal error. ERA5-Land and MSWEP significantly overestimated (2,504.3 mm/a) and underestimated (1,662.2 mm/a) the average-annual precipitation, respectively. (3) Spatial distribution pattern analysis revealed that the observed multi-year annual precipitation in Hainan Island ranges from 996.9 to 2, 368.9 mm, exhibiting an annular-distribution pattern with higher precipitation in the east than in the west and the southwestern mountainous areas than in the northeastern plains. The precipitation range of 1,337.9‒2,287.0 mm observed in GsMAP was the closest to the rain gauge data and particularly matched that of the high-value center in the southeast of the island. (4) In a precipitation trend analysis, CMORPH, ERA5-Land, GPM, MSWEP, CHIRPS, and PERSIANN showed an increasing trend in local areas of Hainan Island, while GsMAP showed a stronger increasing trend. (5) In an analysis of extreme precipitation events, GsMAP, CMORPH, and GPM reproduced the spatiotemporal evolution of extreme precipitation events on a daily scale in Hainan relatively well. GPM better reproduced the spatial and temporal evolution characteristics of typhoon precipitation events in Hainan Island. However, the accuracy of the precipitation estimation still requires improvement. The results of this study not only contribute to our understanding of precipitation products applicable to Hainan but also provide insights for improving satellitebased precipitation products in tropical island environments. These findings underscore the importance of regional validation and the potential of multi-product fusion approaches for enhancing precipitation estimates in complex terrains. [ABSTRACT FROM AUTHOR]

Taking the households of urban and rural residents in Zhejiang Province as the research object in 2020, an indicator system for rural development and urbanization is constructed from the four aspects of ecological environment, public services, grassroots governance and quality of life, using entropy method and Coupling Coordination Degree Model, Theil index and other methods to summarize the urban-rural development experience and problems of different counties and cities in Zhejiang Province, and explore the urban-rural coupling mode and implementation path. The results show that: ①Based on the analysis of different spatial scales, the urbanization and rural have regional imbalances in development, with obvious gradient changes from east to west and little difference between sub-regions. About 80% of the overall difference in urban-rural development in Zhejiang comes from differences within sub-regions, and another 60% comes from the internal differences between southwestern Zhejiang and southern Zhejiang. At the county level, the urbanization and rural development index presents a spatial pattern of "high in the northeast and low in the southwest". ②Based on the analysis of the degree of urban- rural coupling and coordination in Zhejiang Province, four urban-rural coupling modes are summarized: independent parallel coupling, urban-rural partial benefit interactive coupling, urban-rural symmetrical reciprocal coupling, and urban-rural integration symbiosis coupling. ③ Based on the rural revitalization strategy, the implementation path summarized from ecological livability, public services, grassroots governance and quality of life is characterized by ecological integration, green development, spatial integration, shared development, governance integration, good rural governance, factor integration, and common prosperity. [ABSTRACT FROM AUTHOR]

The method of quantum dynamics is employed to investigate the mean strategy in the swarm intelligence algorithm. The physical significance of the population mean point is explained as the location where the optimal solution with the highest likelihood can be found once a quantum system has reached a ground state. Through the use of the double well function and the CEC2013 test suite, controlled experiments are conducted to perform a comprehensive performance analysis of the mean strategy. The empirical results indicate that implementing the mean strategy not only enhances solution diversity but also yields accurate, efficient, stable, and effective outcomes for finding the optimal solution. [ABSTRACT FROM AUTHOR]

Image itself information is naturally robust to image reconstruction, yet most current super-resolution methods do not fully utilize global feature information. This study proposes a new image super-resolution model mixing multiple and multi-scale attentions, including two new modules: Multi-scale hybrid non-local attention upsampling module and residual dense attention block. Different from previous nonlocal methods, multi-scale hybrid non-local attention upsampling module mixes pixel-based and patch-based nonlocal attention and establishes patch-level upsampling mapping relationships at multiple scales, which enables a wider global search space. The residual dense attention block establishes attention associations in channel and spatial dimensions, which enhances the transfer and fusion of front-to-back attention information through dense connections. In this study, quantitative and qualitative evaluations are conducted on several benchmark datasets, and the experimental results show that the model outperforms similar super-resolution models in terms of performance and reconstruction quality. [ABSTRACT FROM AUTHOR]

Road crack detection is of paramount importance for ensuring vehicular traffic safety, and implementing traditional detection methods for cracks inevitably impedes the optimal functioning of traffic. In light of the above, we propose a USSC-YOLO-based target detection algorithm for unmanned aerial vehicle (UAV) road cracks based on machine vision. The algorithm aims to achieve the high-precision detection of road cracks at all scale levels. Compared with the original YOLOv5s, the main improvements to USSC-YOLO are the ShuffleNet V2 block, the coordinate attention (CA) mechanism, and the Swin Transformer. First, to address the problem of large network computational spending, we replace the backbone network of YOLOv5s with ShuffleNet V2 blocks, reducing computational overhead significantly. Next, to reduce the problems caused by the complex background interference, we introduce the CA attention mechanism into the backbone network, which reduces the missed and false detection rate. Finally, we integrate the Swin Transformer block at the end of the neck to enhance the detection accuracy for small target cracks. Experimental results on our self-constructed UAV near–far scene road crack i(UNFSRCI) dataset demonstrate that our model reduces the giga floating-point operations per second (GFLOPs) compared to YOLOv5s while achieving a 6.3% increase in mAP@50 and a 12% improvement in mAP@ [50:95]. This indicates that the model remains lightweight meanwhile providing excellent detection performance. In future work, we will assess road safety conditions based on these detection results to prioritize maintenance sequences for crack targets and facilitate further intelligent management. [ABSTRACT FROM AUTHOR]

Timely discovery and disposal of road risk sources constitute the cornerstone of road operation safety. Presently, the detection of road risk sources frequently relies on manual inspections via inspection vehicles, a process that is both inefficient and time-consuming. To tackle this challenge, this paper introduces a novel automated approach for detecting road risk sources, termed the multi-scale lightweight network (MSLN). This method primarily focuses on identifying road surfaces, potholes, and scattered objects. To mitigate the influence of real-world factors such as noise and uneven brightness on test results, pavement images were carefully collected. Initially, the collected images underwent grayscale processing. Subsequently, the median filtering algorithm was employed to filter out noise interference. Furthermore, adaptive histogram equalization techniques were utilized to enhance the visibility of cracks and the road background. Following these preprocessing steps, the MSLN model was deployed for the detection of road risk sources. Addressing the challenges associated with two-stage network models, such as prolonged training and testing times, as well as deployment difficulties, this study adopted the lightweight feature extraction network MobileNetV2. Additionally, transfer learning was incorporated to elevate the model's training efficiency. Moreover, this paper established a mapping relationship model that transitions from the world coordinate system to the pixel coordinate system. This model enables the calculation of risk source dimensions based on detection outcomes. Experimental results reveal that the MSLN model exhibits a notably faster convergence rate. This enhanced convergence not only boosts training speed but also elevates the precision of risk source detection. Furthermore, the proposed mapping relationship coordinate transformation model proves highly effective in determining the scale of risk sources. [ABSTRACT FROM AUTHOR]

A multi-channel computational ghost imaging (GI) method based on multi-scale speckle optimization is proposed. We not only reduce imaging time and enhance imaging quality but also reduce interference among different channels. Using one bucket detector to receive total light intensity, the color speckle is formed by combining components obtained through the singular value decomposition of three self-designed multi-scale measurement matrices. Simulation and experimental results demonstrate that our designed method contributes to reducing imaging time and enhancing imaging quality, achieving improved visual quality even at low sampling rates. This approach enhances GI flexibility and holds potential for diverse applications, including target recognition and biomedical imaging. [ABSTRACT FROM AUTHOR]

With the development of deep learning, Remote Sensing Image (RSI) semantic segmentation has produced significant advances. However, due to the sparse distribution of the objects and the high similarity between classes, the task of semantic segmentation in RSI is still extremely challenging. In this paper, we propose a novel semantic segmentation framework for RSI called HST-UNet that can overcome the shortcomings of the existing models and extract and recover the global and local features of RSI, which is a hybrid semantic segmentation model with Shunted Transformer as encoder and Multi-Scale Convolutional Attention Network (MSCAN) as decoder. Then, to better fuse the information from the Encoder and the Decoder and alleviate the ambiguity, we design a Learnable Weighted Fusion (LWF) module to effectively connect to the decoder features. Extensive experiments demonstrate that the proposed HST-UNet outperforms the state-of-the-art methods, achieving F1 score/MIoU accuracy of 71.44%/83.00% on the ISPRS Vaihingen dataset and 77.36%/87.09% on ISPRS Potsdam dataset. The code will be available at https://github.com/HC-Zhou/HST-UNet. [ABSTRACT FROM AUTHOR]

Forest fire is a severe natural disaster, which leads to the destruction of forest ecology. At present, fire detection technology represented by convolutional neural network is widely used in forest resource protection, which can realize rapid analysis. However, in forest flame and smoke detection tasks, due to continuous expansion of the target range, a better detection effect cannot be achieved. This paper proposes an improved YOLOX method for multi-scale forest fire detection. This method proposes a novel feature pyramid model to reduce the information loss of high-level forest fire feature maps and enhance the representation ability of feature pyramids. Moreover, the method applies a small object data augmentation strategy to enrich the forest fire dataset, making it more suitable for the actual forest fire scene. According to the experimental results, the mAP of the model proposed in this paper reaches 79.64%, which is about 4.89% higher than the baseline network YOLOX. The method improves the accuracy of forest fire detection, reduces false alarms, and is suitable for real scenarios of forest fires. [ABSTRACT FROM AUTHOR]

Copyright of Journal of South-Central Minzu University (Natural Science Edition) is the property of Journal of South-Central Minzu University (Natural Science Edition) Editorial Office and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Copyright of Journal of Remote Sensing is the property of Editorial Office of Journal of Remote Sensing & Science Publishing Co. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

During magnetospheric substorms, plasma from magnetic reconnection in the magnetotail is thought to reach the inner magnetosphere and form a partial ring current. We simulate this process using a fully kinetic 3D particle‐in‐cell (PIC) numerical code along with a global magnetohydrodynamics (MHD) model. The PIC simulation extends from the solar wind outside the bow shock to beyond the reconnection region in the tail, while the MHD code extends much further and is run for nominal solar wind parameters and a southward interplanetary magnetic field. By the end of the PIC calculation, ions and electrons from the tail reconnection reach the inner magnetosphere and form a partial ring current and diamagnetic current. The primary source of particles to the inner magnetosphere is bursty bulk flows (BBFs) that originate from a complex pattern of reconnection in the near‐Earth magnetotail at xGSM=−18RE ${x}_{\text{GSM}}=-18{R}_{\mathrm{E}}$ to −30RE ${-}30{R}_{\mathrm{E}}$. Most ion acceleration occurs in this region, gaining from 10 to 50 keV as they traverse the sites of active reconnection. Electrons jet away from the reconnection region much faster than the ions, setting up an ambipolar electric field allowing the ions to catch up after approximately 10 ion inertial lengths. The initial energy flux in the BBFs is mainly kinetic energy flux from the ions, but as they move earthward, the energy flux changes to enthalpy flux at the ring current. The power delivered from the tail reconnection in the simulation to the inner magnetosphere is >2×1011 ${ >} 2\times 1{0}^{11}$ W, which is consistent with observations. Plain Language Summary: During intervals of increased solar activity, the magnetic field in Earth's stretched night‐side tail undergoes intense reconfiguration that can energize particles. This process is called magnetic reconnection. Ions and electrons from reconnection can reach the inner magnetosphere. In this paper, we simulate this process using a novel model that includes Earth's global magnetic field configuration and self‐consistently models particle physics for both electrons and ions. We find that the particles are accelerated significantly near the sites of magnetic field reconfiguration with ions gaining 10 s of keV energy. As they propagate earthward, they end up contributing energetically to the formation of a ring current system partially encircling Earth. Key Points: Ions and electrons accelerated by 10–50 keV near the magnetotail reconnection can reach the inner magnetosphereA partial ring current is formed during the simulation, with highest energy flux between midnight and duskThe ion and electron energy fluxes are mostly kinetic in the reconnection region but change to enthalpy flux earthward [ABSTRACT FROM AUTHOR]

Hand gesture segmentation is an important research topic in computer vision. Despite ongoing efforts, achieving optimal gesture segmentation remains challenging, attributed to factors like gesture morphology and intricate backgrounds. In light of these challenges, we propose a novel hand gesture segmentation approach that strategically combines the strengths of Convolutional Neural Networks (CNN) for local feature extraction and Transformer Networks for global feature integration. To be more specific, we design two feature fusion modules. One employs an attention mechanism to learn how to fuse features extracted by CNN and Transformer. The second module utilizes a combination of group convolution and activation functions to implement gating mechanisms, enhancing the response of crucial features while minimizing interference from weaker ones. Our proposed method achieves mIoU score of 93.53%, 97.25%, and 90.39% on OUHANDS, HGR1, and EgoHands hand gesture datasets respectively, which outperforms the state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Computer Engineering & Applications is the property of Beijing Journal of Computer Engineering & Applications Journal Co Ltd. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Copyright of Journal of Xi'an Jiaotong University is the property of Editorial Office of Journal of Xi'an Jiaotong University and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Copyright of Journal of Shenzhen University Science & Engineering is the property of Editorial Department of Journal of Shenzhen University Science & Engineering and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Effective representation of visual feature extraction is the key to improving continuous sign language recognition performance. However, the differences in the temporal length of sign language actions and the sign language weak annotation problem make effective visual feature extraction more difficult. To focus on the above problems, a method named multi-scale visual feature extraction and cross-modality alignment for continuous sign language recognition (MECA) is proposed. The method mainly consists of a multi-scale visual feature extraction module and cross-modal alignment constraints. Specifically, in the multi-scale visual feature extraction module, the bottleneck residual structures with different dilated factors are fused in parallel to enrich the multi-scale temporal receptive field for extracting visual features with different temporal lengths. Furthermore, the hierarchical reuse design is adopted to further strengthen the visual feature. In the cross-modality alignment constraint, dynamic time warping is used to model the intrinsic relationship between sign language visual features and textual features, where textual feature extraction is achieved by the collaboration of a multilayer perceptron and a long short-term memory network. Experiments performed on the challenging public datasets RWTH-2014, RWTH-2014T and CSL-Daily show that the proposed method achieves competitive performance. The results demonstrate that the multi-scale approach proposed in MECA can capture sign language actions of distinct temporal lengths, and constructing the cross-modal alignment constraint is correct and effective for continuous sign language recognition under weak supervision.

Abstract This paper addresses the challenge of low detection accuracy of grape clusters caused by scale differences, illumination changes, and occlusion in realistic and complex scenes. We propose a multi-scale feature fusion and augmentation YOLOv7 network to enhance the detection accuracy of grape clusters across variable environments. First, we design a Multi-Scale Feature Extraction Module (MSFEM) to enhance feature extraction for small-scale targets. Second, we propose the Receptive Field Augmentation Module (RFAM), which uses dilated convolution to expand the receptive field and enhance the detection accuracy for objects of various scales. Third, we present the Spatial Pyramid Pooling Cross Stage Partial Concatenation Faster (SPPCSPCF) module to fuse multi-scale features, improving accuracy and speeding up model training. Finally, we integrate the Residual Global Attention Mechanism (ResGAM) into the network to better focus on crucial regions and features. Experimental results show that our proposed method achieves a mAP $$_{0.5}$$ 0.5 of 93.29% on the GrappoliV2 dataset, an improvement of 5.39% over YOLOv7. Additionally, our method increases Precision, Recall, and F1 score by 2.83%, 3.49%, and 0.07, respectively. Compared to state-of-the-art detection methods, our approach demonstrates superior detection performance and adaptability to various environments for detecting grape clusters.

Global precipitation observations have been realized through the development of satellite remote-sensing technology. However, there is a lack of evaluation of remote-sensing precipitation products in complex tropical island terrains. This study used hourly rain gauge data to conduct a multi-scale systematic evaluation of common precipitation products, such as CMORPH, CHIRPS, GsMAP, GPM, MSWEP, ERA5-Land, and PERSIANN, over Hainan Island, providing an in-depth analysis of the precipitation detection capabilities of various products in this region. The main conclusions are: (1) In a multi-temporal scale evaluation, GPM and GsMAP outperformed the other products across all time scales. On a 3-hour scale, GPM and GsMAP showed the highest correlation coefficients (0.53 and 0.52, respectively). On a daily scale, except for PERSIANN, all products showed correlation coefficients above 0.56, with GPM and GsMAP showing the best performance (R = 0.73 and 0.74, respectively). (2) In comparing annual precipitation, Hainan Island's average-annual precipitation over the past 20 years showed a fluctuating trend, with a mean of 1,776.4 mm/a. The CMORPH annual average of 1,765.1 mm/a was the closest to the CHM-PRE dataset, with minimal error. ERA5-Land and MSWEP significantly overestimated (2,504.3 mm/a) and underestimated (1,662.2 mm/a) the average-annual precipitation, respectively. (3) Spatial distribution pattern analysis revealed that the observed multi-year annual precipitation in Hainan Island ranges from 996.9 to 2,368.9 mm, exhibiting an annular-distribution pattern with higher precipitation in the east than in the west and the southwestern mountainous areas than in the northeastern plains. The precipitation range of 1,337.9‒2,287.0 mm observed in GsMAP was the closest to the rain gauge data and particularly matched that of the high-value center in the southeast of the island. (4) In a precipitation trend analysis, CMORPH, ERA5-Land, GPM, MSWEP, CHIRPS, and PERSIANN showed an increasing trend in local areas of Hainan Island, while GsMAP showed a stronger increasing trend. (5) In an analysis of extreme precipitation events, GsMAP, CMORPH, and GPM reproduced the spatiotemporal evolution of extreme precipitation events on a daily scale in Hainan relatively well. GPM better reproduced the spatial and temporal evolution characteristics of typhoon precipitation events in Hainan Island. However, the accuracy of the precipitation estimation still requires improvement. The results of this study not only contribute to our understanding of precipitation products applicable to Hainan but also provide insights for improving satellite-based precipitation products in tropical island environments. These findings underscore the importance of regional validation and the potential of multi-product fusion approaches for enhancing precipitation estimates in complex terrains.

In recent years, skeleton-based action recognition has received considerable attention due to the robustness of human skeletons in complex environments. However, many existing methods face challenges in effectively learning global temporal information due to inadequate extraction of spatiotemporal features and the neglect of long-term dependencies. Furthermore, subtle joint movements play a critical role in skeleton-based behavior recognition, as such movements are essential for distinguishing between similar actions. To address the aforementioned challenges, this paper proposes a Multi-Scale Spatiotemporal Topology-Aware Network (MSTC3D), which integrates data from various sampled frames into a dual-channel network and employs lateral connections to merge features from different temporal scales. This facilitates the dynamic learning of global temporal channel variations, enhancing the modeling of long-term temporal dependencies. The proposed Multi-Scale 3D Convolutional Block (M3D) incorporates a pyramid-like structure to expand the receptive field effectively, thereby enabling the accurate capture of multi-layered detailed features of subtle joint movements. Moreover, to further enhance the model’s fine-grained recognition capability for features associated with various joints and regions, a Spatial Topological Focus Module is embedded within the M3D. By comprehensively considering both short-term and long-term temporal dependencies, and leveraging the efficient feature representation provided by multi-scale convolutional blocks, MSTC3D demonstrates superior performance in action recognition tasks. Experiments on the NTU RGB+D and FineGym datasets validate the effectiveness of MSTC3D, showing state-of-the-art performance compared to CNN-based methods and achieving comparable superior performance to leading GCN-based methods. [ABSTRACT FROM AUTHOR]

Shale is a complex porous medium with multi-scale pores and well-developed fracture networks. This paper aims to use modeling and numerical simulation methods to study the transport of shale gas in a complex multi-scale matrix-fracture system. In this study, mathematical modeling and programming was used to establish digital models of shale gas multi-scale matrix and 3D discrete fracture network. Based on the transport mechanisms of shale gas, this study derives the mathematical models of shale gas transport in different transport media, and uses the finite element method to solve and analysis the transport of shale gas in the multi-scale matrix-fracture system. The model is verified by real shale gas field data. The results show that the fractal dimension of organic pores have great effects on shale gas transport. When the fractal dimension is greater than 1.4, the increase on gas production is particularly obvious. Compared with organic pores, the effect of fractal dimension of inorganic pores is smaller. The existence of the fracture network has an effect of up to 25% on gas production, and the optimal fracture density is 200. [ABSTRACT FROM AUTHOR]

Objective: Illumination of the integrative effects of herbs in a formula is a bottleneck that limits the development of traditional Chinese medicine (TCM). In the present study, we developed a transcriptome-based multi-scale network pharmacology model to explore the combined effects of different herbs. Materials and Methods: First, we curated gene signatures at different biological scales, from the molecular to higher tissue levels, including tissues, cells, pathological processes, biological processes, pathways, and targets. Second, using the Xiexin Tang (XXT) formula as an example, we collected transcriptomic data in response to the treatment of XXT or its three compositive herbs on Michigan cancer foundation7 cells. Third, we linked each herbal drug to different biological scales by calculating the correlation scores between herb-induced gene expression profiles and gene signatures. Finally, the combined mechanisms of the three constituent herbs in XXT were deciphered by comparing their multi-scale effects with those of the formula. Results: The results showed that although XXT or single herbs regulated a large number of signatures on each biological scale, the phenotypic effects of these herbal drugs are concentrated onto the “Blood” tissue, types of hemocytes, and hemorrhagic injury-related pathological processes. At the molecular level, these herbs consistently regulate processes such as the cell cycle and blood coagulation-related pathways, as well as protein targets related to the immunoinflammatory response and blood coagulation, such as proteinase-activated receptor 2, integrin beta-3, inhibitor of nuclear factor kappa-B kinase subunit beta, and coagulation factor XII. The analysis of the combinational modes demonstrated that different herbs can cooperate by acting on the same objects and/or regulating different objects in related functions, and cooperative behaviors change at different biological scales. Conclusions: Our model can dissect the combined effects of herbal formulae from a multi-scale perspective and should be beneficial for the development and exploitation of TCM.

Abstract The diagnosis of early prostate cancer depends on the accurate segmentation of prostate regions in magnetic resonance imaging (MRI). However, this segmentation task is challenging due to the particularities of prostate MR images themselves and the limitations of existing methods. To address these issues, we propose a U-shaped encoder-decoder network MM-UNet based on Mamba and CNN for prostate segmentation in MR images. Specifically, we first proposed an adaptive feature fusion module based on channel attention guidance to achieve effective fusion between adjacent hierarchical features and suppress the interference of background noise. Secondly, we propose a global context-aware module based on Mamba, which has strong long-range modeling capabilities and linear complexity, to capture global context information in images. Finally, we propose a multi-scale anisotropic convolution module based on the principle of parallel multi-scale anisotropic convolution blocks and 3D convolution decomposition. Experimental results on two public prostate MR image segmentation datasets demonstrate that the proposed method outperforms competing models in terms of prostate segmentation performance and achieves state-of-the-art performance. In future work, we intend to enhance the model's robustness and extend its applicability to additional medical image segmentation tasks.

Ultra-high performance concrete (UHPC), as one of the most eye-catching building materials, has been the subject of extensive research by scholars. On this basis, to expand the application of UHPC for offshore wind turbine towers in complex marine environments, three different fiber materials - copper-plated microfibre steel fibers, carbon fibers, and carbon nanotubes (CNTs) - have been selected for the study of the possibilities of further improving the mechanical properties of UHPC. This study focused on understanding the impact of various fiber combinations and dosages on the flowability, compressive strength, flexural strength, and tensile strength of UHPC. Our findings indicate that carbon fiber, when present at a concentration of up to 0.5%, the effect on the fluidity of UHPC is only about 1.05%. However, the addition of CNTs significantly diminishes the flowability of UHPC, with a consistent decrease observed as the CNT content increases. Notably, when carbon fiber and CNTs are used in combination, the maximum reduction in flowability reaches 7.8%. Furthermore, as the dosage of these fibers increases, the compressive strength, flexural strength, and tensile strength of UHPC all demonstrate a positive trend of improvement. It is observed that the optimal performance is achieved when both carbon fiber and CNTs are present. In particular, carbon fiber exhibits a more profound impact on enhancing compressive strength and flexural strength, when carbon fibers were doped by volume at 0.5%, the compressive and flexural strengths were increased by 6.7% and 11.7%, respectively, compared to the control group, while carbon nanotubes increased the tensile strength by 7.4% at lower dosage. These findings highlight the potential of fiber combinations to optimize UHPC’s mechanical properties for various engineering applications..