Scale Variations Research Articles

SSCFormer: Revisiting ConvNet-Transformer Hybrid Framework From Scale-Wise and Spatial-Channel-Aware Perspectives for Volumetric Medical Image Segmentation.

Accurate and robust medical image segmentation is crucial for assisting disease diagnosis, making treatment plan, and monitoring disease progression. Adaptive to different scale variations and regions of interest is essential for high accuracy in automatic segmentation methods. Existing methods based on the U-shaped architecture respectively tackling intra- and inter-scale problem with a hierarchical encoder, however, are restricted by the scope of multi-scale modeling. In addition, global attention and scaling attention in regions of interest have not been appropriately adopted, especially for the salient features. To address these two issues, we propose a ConvNet-Transformer hybrid framework named SSCFormer for accurate and versatile medical image segmentation. The intra-scale ResInception and inter-scale transformer bridge are designed to collaboratively capture the intra- and inter-scale features, facilitating the interaction of small-scale disparity information at a single stage with large-scale from multiple stages. Global attention and scaling attention are cleverly integrated from a spatial-channel-aware perspective. The proposed SSCFormer is tested on four different medical image segmentation tasks. Comprehensive experimental results show that SSCFormer outperforms the current state-of-the-art methods.

Precision mapping of snail habitat in lake and marshland areas: Integrating environmental and textural indicators using Random Forest modeling

Schistosomiasis japonica continues to pose a significant public health challenge in China, primarily due to the widespread distribution of Oncomelania hupensis, the sole intermediate host of Schistosoma. This study aims to address the constraints of existing remote sensing analyses for identifying snail habitats, which frequently neglect spatial scale and seasonal variations. To this end, we adopt a multi-source data-driven Random Forest approach that integrates bottomland and ground-surface texture data with traditional environmental variables, enhancing the accuracy of snail habitat assessments. We developed four distinct models for the lake and marshland areas of Guichi, China: a baseline model incorporating ground-surface texture, bottomland variables, and environmental variables; Model 1 with only environmental variables; Model 2 adding ground-surface texture and environmental variables; and Model 3 integrating bottomland with environmental variables. The baseline model outperformed the others, achieving a true skill statistic of 0.93, an accuracy of 0.97, a kappa statistic of 0.94, and an area under the curve of 0.99. Our analysis pinpointed critical high-risk snail habitats distributed in a belt-like pattern along major water bodies, near the Yangtze River, QiuPu River, and around Shengjin Lake, Jiuhua River, and Qingtong River. These insights can aid local health authorities in more efficiently allocating limited resources, developing effective snail surveillance and control strategies to combat schistosomiasis. Additionally, this approach can be adapted to localize other endemic hosts with similar ecological characteristics.

Nighttime wildlife object detection based on YOLOv8‐night

AbstractMonitoring nocturnal animals in the field is an important task in ecological research and wildlife conservation, but the complexity of nocturnal images and low light conditions make it difficult to cope with traditional image processing methods. To address this problem, researchers have introduced infrared cameras to improve the accuracy of nocturnal animal behaviour observations. Object detection in nighttime images captured by infrared cameras faces several challenges, including low image quality, animal scale variations, occlusion, and pose changes. This study proposes the YOLOv8‐night model, which effectively overcomes these challenges by introducing a channel attention mechanism in YOLOv8. The model is more focused on capturing animal‐related features by dynamically adjusting the channel weights, which improves the saliency of key features and increases the accuracy rate in complex backgrounds. The main contribution of this study is the introduction of the channel attention mechanism into the YOLOv8 framework to create a YOLOv8‐night model suitable for object detection in nighttime images. When tested on nighttime images, the model performs well with a significantly higher mAP (0.854) than YOLOv8 (0.831), and YOLOv8‐night scores 0.856 on mAP_l, which is obviously better than YOLOv8 (0.833) in terms of processing large objects. The study provides a reliable technical tool for ecological research, wildlife conservation and environmental monitoring, and offers new methods and insights for the study of nocturnal animal behaviour.

Exploring galaxy properties of eCALIFA with contrastive learning

Contrastive learning (CL) has emerged as a potent tool for building meaningful latent representations of galaxy properties across a broad spectrum of wavelengths, ranging from optical and infrared to radio frequencies. These latent representations facilitate a variety of downstream tasks, including galaxy classification, similarity searches in extensive datasets, and parameter estimation, which is why they are often referred to as foundation models for galaxies. In this study, we employ CL on the latest extended data release from the Calar Alto Legacy Integral Field Area (CALIFA) survey, which encompasses a total of 895 galaxies with enhanced spatial resolution that reaches the limits imposed by natural seeing (FWHMPSF ∼ 1.5). We demonstrate that CL can be effectively applied to Integral Field Unit (IFU) surveys, even with relatively small training sets, to construct meaningful embedding where galaxies are well separated based on their physical properties. We discover that the strongest correlations in the embedding space are observed with the equivalent width of Hα, galaxy morphology, stellar metallicity, luminosity-weighted age, stellar surface mass density, the [NII]/Hα ratio, and stellar mass, in descending order of correlation strength. Additionally, we illustrate the feasibility of unsupervised separation of galaxy populations along the star formation main sequence, successfully identifying the blue cloud and the red sequence in a two-cluster scenario, and the green valley population in a three-cluster scenario. Our findings indicate that galaxy luminosity profiles have minimal impact on the construction of the embedding space, suggesting that morphology and spectral features play a more significant role in distinguishing between galaxy populations. Moreover, we explore the use of CL for detecting variations in galaxy population distributions across different large-scale structures, including voids, clusters, and filaments and walls. Nonetheless, we acknowledge the limitations of the CL framework and our specific training set in detecting subtle differences in galaxy properties, such as the presence of an AGN or other minor scale variations that exceed the scope of primary parameters such as the stellar mass or morphology. Conclusively, we propose that CL can serve as an embedding function for the development of larger models capable of integrating data from multiple datasets, thereby advancing the construction of more comprehensive foundation models for galaxies.

Population shrinkage and its influence on housing prices: A mediation perspective based on commercial amenities

The effects of population shrinkage have become a new topic of concern. Existing research has explored the effects of population shrinkage on housing prices within specific cities, yet due to scale variations, the effects of population shrinkage on housing prices may exhibit uncertainty, are conclusions drawn at the micro level applicable to the macro level? Currently, there is limited exploration of the collective effects of micro mechanisms of population shrinkage on housing prices. In this paper, we outline the effects of population shrinkage on housing prices at both macro and micro levels. Subsequently, we propose to explore the general effects of population shrinkage on housing prices based on commercial amenities. Specifically, we analyze the spatial patterns and characteristics of population shrinkage and housing prices in 1235 Chinese shrinking counties. Furthermore, using partial least squares structural equation modeling (PLS-SEM), we examine how commercial amenities mediate the relationship between population shrinkage and housing prices. This study contributes theoretically and empirically to understanding the impact of population shrinkage on housing prices, while also holds practical significance for regulating the effects of population shrinkage and stabilizing the value of real estate markets from the perspective of commercial amenities.

Investigating the effect of scaling and temperature on the performance of improved junctionless nanowire FET through simulation analysis

An Improved Junctionless Nanowire Field Effect Transistor (I-JL-NWFET) device is proposed in this paper to address the limitations of conventional JL-NWFET. This research paper initially, comprehensively analyzes the impact of channel length (L) and channel thickness (t si ) scaling on the electrical, analog/RF, and linearity performance of I-JL-NWFET and JL-NWFET. The results suggest that the specific design features in I-JL-NWFET contribute to a more robust and less sensitive response to variations in scaling compared to its counterpart, JL-NWFET. Furthermore, an exploration into the impact of temperature on the electrical, analog/RF, and linearity performance is also conducted for both I-JL-NWFET and JL-NWFET. The electrical performance of I-JL-NWFET showcases a significantly reduced temperature sensitivity in parameters like drain current (I D ), Subthreshold Slope (SS) and Drain Induced Barrier Lowering (DIBL) compared to JL-NWFET. Subsequently, analyzing the analog/RF performance in the context of parameters such as transconductance (g m ), Transconductance Gain Factor (TGF), output conductance (g d ), early voltage (V EA ), total gate capacitance (C GG ), and cut-off frequency (f T ) under temperature variation, a lower degree of variability in I-JL-NWFET is observed compared to JL-NWFET. Furthermore, the linearity performance of I-JL-NWFET, assessed through parameters such as second and third-order transconductance (g m2 , g m3 ), second and third-order input voltage intercept points (VIP2, VIP3), and third-order intermodulation distortion (IIP3 and IMD3) is improved at the higher temperature than that of JL-NWFET.

Multiple Object Tracking in Drone Aerial Videos by a Holistic Transformer and Multiple Feature Trajectory Matching Pattern

Drone aerial videos have immense potential in surveillance, rescue, agriculture, and urban planning. However, accurately tracking multiple objects in drone aerial videos faces challenges like occlusion, scale variations, and rapid motion. Current joint detection and tracking methods often compromise accuracy. We propose a drone multiple object tracking algorithm based on a holistic transformer and multiple feature trajectory matching pattern to overcome these challenges. The holistic transformer captures local and global interaction information, providing precise detection and appearance features for tracking. The tracker includes three components: preprocessing, trajectory prediction, and matching. Preprocessing categorizes detection boxes based on scores, with each category adopting specific matching rules. Trajectory prediction employs the visual Gaussian mixture probability hypothesis density method to integrate visual detection results to forecast object motion accurately. The multiple feature pattern introduces Gaussian, Appearance, and Optimal subpattern assignment distances for different detection box types (GAO trajectory matching pattern) in the data association process, enhancing tracking robustness. We perform comparative validations on the vision-meets-drone (VisDrone) and the unmanned aerial vehicle benchmarks; the object detection and tracking (UAVDT) datasets affirm the algorithm’s effectiveness: it obtained 38.8% and 61.7% MOTA, respectively. Its potential for seamless integration into practical engineering applications offers enhanced situational awareness and operational efficiency in drone-based missions.

Recent Advances for Aerial Object Detection: A Survey

Aerial object detection, as object detection in aerial images captured from an overhead perspective, has been widely applied in urban management, industrial inspection, and other aspects. However, the performance of existing aerial object detection algorithms is hindered by variations in object scales and orientations attributed to the aerial perspective. This survey presents a comprehensive review of recent advances in aerial object detection. We start with some basic concepts of aerial object detection and then summarize the five imbalance problems of aerial object detection, including scale imbalance, spatial imbalance, objective imbalance, semantic imbalance, and class imbalance. Moreover, we classify and analyze relevant methods and especially introduce the applications of aerial object detection in practical scenarios. Finally, the performance evaluation is presented on two popular aerial object detection datasets VisDrone-DET and DOTA, and we discuss several future directions that could facilitate the development of aerial object detection.

Contextual Enhancement–Interaction and Multi-Scale Weighted Fusion Network for Aerial Tracking

Siamese-based trackers have been widely utilized in UAV visual tracking due to their outstanding performance. However, UAV visual tracking encounters numerous challenges, such as similar targets, scale variations, and background clutter. Existing Siamese trackers face two significant issues: firstly, they rely on single-branch features, limiting their ability to achieve long-term and accurate aerial tracking. Secondly, current tracking algorithms treat multi-level similarity responses equally, making it difficult to ensure tracking accuracy in complex airborne environments. To tackle these challenges, we propose a novel UAV tracking Siamese network named the contextual enhancement–interaction and multi-scale weighted fusion network, which is designed to improve aerial tracking performance. Firstly, we designed a contextual enhancement–interaction module to improve feature representation. This module effectively facilitates the interaction between the template and search branches and strengthens the features of each branch in parallel. Specifically, a cross-attention mechanism within the module integrates the branch information effectively. The parallel Transformer-based enhancement structure improves the feature saliency significantly. Additionally, we designed an efficient multi-scale weighted fusion module that adaptively weights the correlation response maps across different feature scales. This module fully utilizes the global similarity response between the template and the search area, enhancing feature distinctiveness and improving tracking results. We conducted experiments using several state-of-the-art trackers on aerial tracking benchmarks, including DTB70, UAV123, UAV20L, and UAV123@10fps, to validate the efficacy of the proposed network. The experimental results demonstrate that our tracker performs effectively in complex aerial tracking scenarios and competes well with state-of-the-art trackers.

DSNet: dual-stream network for fine-grained ship classification in optical remote sensing images

ABSTRACT The fine-grained ship classification (FGSC) task in remote sensing aims to distinguish subordinate ship categories. Due to the high similarity among different subcategories and the significant variations in pose, scale, rotation, and color within the same subcategory in the fine-grained ship datasets, the FGSC task is extremely challenging. To alleviate these issues, a novel approach, named Dual-Stream Network (DSNet), is proposed in this paper for FGSC. DSNet consists of two sub-branches: the Local Residual Branch (LRB) and the Global Representation Branch (GRB). Specifically, the LRB effectively extracts the most discriminative local fine-grained features of the target using Full-scale Skip Connections (FSC) and Stripe Pooling Add (SPA) modules. The GRB captures global contextual information through Swin-Transformer (Swin-T), establishing remote dependencies of the target and extracting global coarse-grained features. Furthermore, a Lightweight Fusion (LF) block is designed in the feature fusion module to combine the feature information from the two branches. This ensures that the discriminative global and local information complement each other without overwhelming one another, thereby effectively improving the performance of the model. Extensive experiments and analyses demonstrate that our proposed method achieves state-of-the-art performance on two mainstream FGSC datasets in remote sensing.

Attention mechanism based on deep learning for defect detection of wind turbine blade via multi-scale features

Abstract An enhanced wind turbine blade surface defect detection algorithm, CGIW-YOLOv8, has been introduced to tackle the problems of uneven distribution of defect samples, confusion between defects and background, and variations in target scales that arise during drone maintenance of wind turbine blades. This algorithm is given based on the YOLOv8 model. Initially, a data augmentation method based on geometric changes and Poisson mixing was used to enrich the dataset and address the problem of uneven sample distribution. Subsequently, the incorporation of the Coordinate Attention (CA) mechanism into the Backbone network improved the feature extraction capability in complex backgrounds. In the Neck, the Reparameterized Generalized Feature Pyramid Network (Rep-GFPN) was introduced as a path fusion strategy and multiple cross-scale connections are fused, which effectively enhances the multi-scale expression ability of the network. Finally, the original CIOU loss function was replaced with Inner-WIoU, which was created by applying the Inner-IoU loss function to the Wise-IoU loss function. It improved detection accuracy while simultaneously speeding up the model's rate of convergence. Experimental results show that the mAP of the method for defect detection reaches 92%, which is 5.5% higher than the baseline network. The detection speed is 120.5 FPS, which meets the needs of real-time detection.

Comparison of Analgesic Method in Laparoscopic Gastrectomy Using External Oblique Intercostal Block Versus Wound Infiltration: A Randomized Controlled Trial.

Background: External oblique intercostal block (EOIB) is effective in relieving pain in the T6 to T10 dermatomes; however, there is limited evidence comparing EOIB with conventional regional anesthesia. In this randomized controlled trial, it was hypothesized that EOIB would provide more effective analgesia than wound infiltration (WI) in laparoscopic gastrectomy. Methods: Thirty-two patients (aged 20-85 years) with American Society of Anesthesiologists Performance Status of 1 to 2, scheduled for laparoscopic gastrectomy, were randomly divided into EOIB and WI groups. Both procedures were performed using 40 mL of 0.25% levobupivacaine. The primary outcome was the numerical rating scale (NRS) score 12 h postoperatively. Secondary outcomes were NRS at 2, 24, and 48 h postoperatively, variation in quality of recovery-15 score on postoperative days 1 and 2 from preoperative baseline, postoperative fentanyl consumption, and variation in World Health Organization Disability Assessment Scale 2.0 scores between baseline and 3 months postoperatively. Plasma concentrations of levobupivacaine and pinprick sensation in the T4-11 dermatomes post-EOIB were evaluated to determine the exploratory endpoints. Results: There were no differences in the NRS scores 12 h postoperatively at rest and during movement between the EOIB and WI groups (mean standard deviation 1.3 [1.1] vs. 1.5 [1.4] and 3.1 [1.6] vs. 3.8 [1.7], respectively). Secondary outcomes did not differ significantly between the groups. The time to peak plasma concentration of levobupivacaine following EOIB was 45 min. Conclusions: No significant differences in NRS scores were observed between the analgesic effects of EOIB and WI at 12 h postoperatively.

E-BDL: Enhanced Band-Dependent Learning Framework for Augmented Radar Sensing.

Radar sensors, leveraging the Doppler effect, enable the nonintrusive capture of kinetic and physiological motions while preserving privacy. Deep learning (DL) facilitates radar sensing for healthcare applications such as gait recognition and vital-sign measurement. However, band-dependent patterns, indicating variations in patterns and power scales associated with frequencies in time-frequency representation (TFR), challenge radar sensing applications using DL. Frequency-dependent characteristics and features with lower power scales may be overlooked during representation learning. This paper proposes an Enhanced Band-Dependent Learning framework (E-BDL) comprising an adaptive sub-band filtering module, a representation learning module, and a sub-view contrastive module to fully detect band-dependent features in sub-frequency bands and leverage them for classification. Experimental validation is conducted on two radar datasets, including gait abnormality recognition for Alzheimer's disease (AD) and AD-related dementia (ADRD) risk evaluation and vital-sign monitoring for hemodynamics scenario classification. For hemodynamics scenario classification, E-BDL-ResNet achieves competitive performance in overall accuracy and class-wise evaluations compared to recent methods. For ADRD risk evaluation, the results demonstrate E-BDL-ResNet's superior performance across all candidate models, highlighting its potential as a clinical tool. E-BDL effectively detects salient sub-bands in TFRs, enhancing representation learning and improving the performance and interpretability of DL-based models.

A Remote Sensing Image Super-Resolution Reconstruction Model Combining Multiple Attention Mechanisms.

Remote sensing images are characterized by high complexity, significant scale variations, and abundant details, which present challenges for existing deep learning-based super-resolution reconstruction methods. These algorithms often exhibit limited convolutional receptive fields and thus struggle to establish global contextual information, which can lead to an inadequate utilization of both global and local details and limited generalization capabilities. To address these issues, this study introduces a novel multi-branch residual hybrid attention block (MBRHAB). This innovative approach is part of a proposed super-resolution reconstruction model for remote sensing data, which incorporates various attention mechanisms to enhance performance. First, the model employs window-based multi-head self-attention to model long-range dependencies in images. A multi-branch convolution module (MBCM) is then constructed to enhance the convolutional receptive field for improved representation of global information. Convolutional attention is subsequently combined across channels and spatial dimensions to strengthen associations between different features and areas containing crucial details, thereby augmenting local semantic information. Finally, the model adopts a parallel design to enhance computational efficiency. Generalization performance was assessed using a cross-dataset approach involving two training datasets (NWPU-RESISC45 and PatternNet) and a third test dataset (UCMerced-LandUse). Experimental results confirmed that the proposed method surpassed the existing super-resolution algorithms, including Bicubic interpolation, SRCNN, ESRGAN, Real-ESRGAN, IRN, and DSSR in the metrics of PSNR and SSIM across various magnifications scales.

Scaling variation in the pinch-off of colloid-polymer mixtures

Hypothesis: The scaling laws of drop pinch-off are known to be affected by drop compositions including dissolved polymers and non-Brownian particles. When the size of the particles is comparable to the characteristic length scale of the polymer network, these particles may interact strongly with the polymer environment, leading to new types of scaling behaviors not reported before.Experiments: Using high-speed imaging, we experimentally studied the time evolution of the neck diameter hmin of drops composed of silica nanoparticles dispersed in PEO solution when extruded from a nozzle.Findings: After initial Newtonian necking with hmin ∼ t2/3, the subsequent stage may exhibit scaling variation, characterized by either exponential or power-law decay, depending on the nanoparticle volume fraction ϕ. The exponential decay hmin ∼ e-t/τ signifies the coil-stretch transition in typical viscoelastic suspensions. We conducted an analysis of the power-law scenario hmin ∼ tα at high ϕ, categorizing the entire process into three distinct regimes based on the exponents α. The dependences of critical thicknesses at transition points and exponents on polymer concentration offer initial insights into the potential transition from heterogeneous to homogeneous thinning in the mixture. This novel scaling variation bears implications for accurately predicting and controlling droplet fragmentation in industrial applications.

Decoupling of carbon burial from productivity in the northeast Indian Ocean

The concentration of atmospheric carbon dioxide (CO2) is a crucial climate parameter as it has far-reaching implications on global temperature. The oceans are a significant sink for CO2. Biologically mediated carbon sequestration, in the form of both inorganic (CaCO3) and organic carbon (Corg), and its subsequent burial in marine sediments play a vital role in regulating atmospheric CO2. Understanding the distribution of carbon in marine sediments under different environments can help predict the fate of excess CO2 in the future. We studied the factors affecting the basin scale variation in carbon burial in the climatically sensitive northeast Indian Ocean, by using the data [CaCO3, Corg, Corg/Nitrogen, and isotopic ratio (δ13C, δ15N) of organic carbon] from a total of 718 surface sediments. The entire continental shelf and slope contain <10 % CaCO3. The highest CaCO3 is in the deepest parts of the central northeast Indian Ocean, away from the mouth of major river systems. Despite of the high productivity, the low Corg on the continental shelf is attributed to the well-oxygenated coarse-grained sediments. The lowest Corg is found in the well-oxygenated deeper central northeast Indian Ocean. Interestingly, the highest total carbon is in the deeper central and equatorial regions, far away from the highly productive marginal marine regions. Our study reveals that the grain size, terrigenous dilution, dissolved oxygen, and water masses strongly influence carbon accumulation in the northeast Indian Ocean, with only secondary influence of the productivity.

Enhancing microwave dielectric properties of Li4NbO4F for LTCC applications through tantalum ion doping

The atomic scale variation of tantalum ion doped Li4NbO4F ceramics is initially predicted through first principle calculations. Subsequently, Li4Nb(1-x)TaxO4F (0.00 ≤ x ≤ 0.10) ceramics are prepared using the solid phase reaction method. After sintering at 875 °C for 4 h, the Li4Nb0.96Ta0.04O4F ceramics exhibit optimum dielectric properties: εr = 14.73 ± 0.19, Q × f = 83,439 ± 2810 GHz (@9.31 GHz, improved by 41 %), τf = - 28.33 ± 1.36 ppm/°C. Both the increase in relative density from 95.9 ± 0.32 % to 96.2 ± 0.34 % and the increase in grain size from 14.9 μm to 15.6 μm are non-intrinsic factors that contribute to the optimization of the dielectric properties. Instead, the internal factors suggest that the primary cause of the gradual decrease in εr is the redshift of the Raman wave number. The reduction in dielectric loss is attributable to the weakening of damping behavior, as well as the increase in lattice energy and electron cloud density. Finally, promising potential of Li4Nb0.96Ta0.04O4F ceramics for low-temperature co-fired ceramic (LTCC) applications is indicated by their favourable chemical compatibility with Ag electrodes.

ICDW-YOLO: An Efficient Timber Construction Crack Detection Algorithm.

A robust wood material crack detection algorithm, sensitive to small targets, is indispensable for production and building protection. However, the precise identification and localization of cracks in wooden materials present challenges owing to significant scale variations among cracks and the irregular quality of existing data. In response, we propose a crack detection algorithm tailored to wooden materials, leveraging advancements in the YOLOv8 model, named ICDW-YOLO (improved crack detection for wooden material-YOLO). The ICDW-YOLO model introduces novel designs for the neck network and layer structure, along with an anchor algorithm, which features a dual-layer attention mechanism and dynamic gradient gain characteristics to optimize and enhance the original model. Initially, a new layer structure was crafted using GSConv and GS bottleneck, improving the model's recognition accuracy by maximizing the preservation of hidden channel connections. Subsequently, enhancements to the network are achieved through the gather-distribute mechanism, aimed at augmenting the fusion capability of multi-scale features and introducing a higher-resolution input layer to enhance small target recognition. Empirical results obtained from a customized wooden material crack detection dataset demonstrate the efficacy of the proposed ICDW-YOLO algorithm in effectively detecting targets. Without significant augmentation in model complexity, the mAP50-95 metric attains 79.018%, marking a 1.869% improvement over YOLOv8. Further validation of our algorithm's effectiveness is conducted through experiments on fire and smoke detection datasets, aerial remote sensing image datasets, and the coco128 dataset. The results showcase that ICDW-YOLO achieves a mAP50 of 69.226% and a mAP50-95 of 44.210%, indicating robust generalization and competitiveness vis-à-vis state-of-the-art detectors.

Rapid predictions of pressure and velocity fields in a 5 × 5 bending fuel assembly based on a POD-based black-box model

The fuel assembly is crucial in pressurized water reactors (PWRs) as it can undergo bending deformation due to flow-induced vibrations (FIVs), impacting its flow characteristics. To evaluate the effect of different bending scales on flow behaviors, the Computational Fluid Dynamics (CFD) model of 5 × 5 fuel assembly is developed. This configuration considers common beam-like deformations: 1st-order C-shape, 2nd-order S-shape, and 3rd-order W-shape. The flow characteristics of the 5 × 5 bending fuel assembly with bending scales of 1 mm and 2 mm are studied, aiming to explore how variations in deformation scales impact these flow characteristics. Subsequently, CFD simulations are conducted for various flow velocities, bending shapes, and scales, with the simulation results organized into a snapshot matrix. Through the Proper Orthogonal Decomposition (POD) method, both the POD coefficients and modes of the snapshot matrix are determined. A black-box model is subsequently utilized to predict the POD coefficients of non-sample cases. The research enables rapid predictions of pressure and velocity fields within a 5 × 5 fuel assembly for any given flow velocity, bending shape, and scale by integrating the predicted POD coefficients and modes. Relative to comprehensive CFD simulations, the POD-based black-box model presented in this study provides a rapid and precise approach for pressure and velocity fields prediction, demonstrating its effectiveness and accuracy.

Shape-Scale Co-Awareness Network for 3D Brain Tumor Segmentation.

The accurate segmentation of brain tumor is significant in clinical practice. Convolutional Neural Network (CNN)-based methods have made great progress in brain tumor segmentation due to powerful local modeling ability. However, brain tumors are frequently pattern-agnostic, i.e. variable in shape, size and location, which can not be effectively matched by traditional CNN-based methods with local and regular receptive fields. To address the above issues, we propose a shape-scale co-awareness network (S2CA-Net) for brain tumor segmentation, which can efficiently learn shape-aware and scale-aware features simultaneously to enhance pattern-agnostic representations. Primarily, three key components are proposed to accomplish the co-awareness of shape and scale. The Local-Global Scale Mixer (LGSM) decouples the extraction of local and global context by adopting the CNN-Former parallel structure, which contributes to obtaining finer hierarchical features. The Multi-level Context Aggregator (MCA) enriches the scale diversity of input patches by modeling global features across multiple receptive fields. The Multi-Scale Attentive Deformable Convolution (MS-ADC) learns the target deformation based on the multiscale inputs, which motivates the network to enforce feature constraints both in terms of scale and shape for optimal feature matching. Overall, LGSM and MCA focus on enhancing the scale-awareness of the network to cope with the size and location variations, while MS-ADC focuses on capturing deformation information for optimal shape matching. Finally, their effective integration prompts the network to perceive variations in shape and scale simultaneously, which can robustly tackle the variations in patterns of brain tumors. The experimental results on BraTS 2019, BraTS 2020, MSD BTS Task and BraTS2023-MEN show that S2CA-Net has superior overall performance in accuracy and efficiency compared to other state-of-the-art methods. Code: https://github.com/jiangyu945/S2CA-Net.