Axial Module Research Articles

An unpaired SAR-to-optical image translation method based on Schrödinger bridge network and multi-scale feature fusion

SAR-to-optical (S2O) translation is able to covert SAR into optical images, which help the interpreter to extract information efficiently. In the absence of strictly matched datasets, it is difficult for existing methods to complete training on unpaired data with a minimum amount of data. By employing the recent Schrödinger bridge-based transformation framework, a multiscale axial residual module (MARM) based on the concept of multi-scale feature fusion has been proposed in this paper. To enable efficient translation of SAR to optical images, the generator and discriminator of the model have been designed. Extensive experiments on the SEN1-2 dataset conducted, and the results show the superiority of the proposed method in terms of the generation quality. Compared with the classical CycleGAN, the proposed method can improve the FID metrics by 42.05%.

Enhancing Proton Radiosensitivity of Chondrosarcoma Using Nanoparticle-Based Drug Delivery Approaches: A Comparative Study of High- and Low-Energy Protons.

To overcome chondrosarcoma's (CHS) high chemo- and radioresistance, we used polyethylene glycol-encapsulated iron oxide nanoparticles (IONPs) for the controlled delivery of the chemotherapeutic doxorubicin (IONPDOX) to amplify the cytotoxicity of proton radiation therapy. Human 2D CHS SW1353 cells were treated with protons (linear energy transfer (LET): 1.6 and 12.6 keV/µm) with and without IONPDOX. Cell survival was assayed using a clonogenic test, and genotoxicity was tested through the formation of micronuclei (MN) and γH2AX foci, respectively. Morphology together with spectral fingerprints of nuclei were measured using enhanced dark-field microscopy (EDFM) assembled with a hyperspectral imaging (HI) module and an axial scanning fluorescence module, as well as scanning electron microscopy (SEM) coupled with energy-dispersive X-Ray spectroscopy (EDX). Cell survival was also determined in 3D SW3153 spheroids following treatment with low-LET protons with/without the IONPDOX compound. IONPDOX increased radiosensitivity following proton irradiation at both LETs in correlation with DNA damage expressed as MN or γH2AX. The IONPDOX-low-LET proton combination caused a more lethal effect compared to IONPDOX-high-LET protons. CHS cell biological alterations were reflected by the modifications in the hyperspectral images and spectral profiles, emphasizing new possible spectroscopic markers of cancer therapy effects. Our findings show that the proposed treatment combination has the potential to improve the management of CHS.

Research on Highly Reliable Self-Powered Vibration Sensors for Geological Drilling

Vibration signals at the bottom of the drill string during geological drilling are crucial for lithological identification and drilling parameter optimization. However, existing downhole vibration sensors suffer from limitations in power supply and reliability. This study proposes a self-powered vibration sensor with high redundancy based on the triboelectric nanogenerator principle, which is capable of measuring both axial and transverse vibrations, thereby reducing the dependence on external power sources. The experimental results show that the sensor can measure axial vibration frequencies ranging from 0 to 11 Hz with an error of less than 4% and transverse vibration frequencies ranging from 0 to 5 Hz with an error of less than 5%. It can operate stably in temperatures from 0 to 180 °C and relative humidities from 0 to 95%. The sensor’s axial vibration measurement features six identical measurement structures, providing high redundancy and effectively enhancing its reliability. Furthermore, the sensor exhibits power generation capabilities. When an external load of 1 MΩ is applied to the axial measurement module and 10 MΩ to the transverse measurement module, the sensor achieves its maximum power output for both axial and transverse measurements, reaching 32.4 × 10−9 W and 2.1 × 10−9 W, respectively. Compared to traditional bottom-of-the-hole vibration sensors, this sensor possesses self-powering capabilities and high reliability, which can improve the operational efficiency and hold significant practical value for future applications.

3D Head Pose Estimation via Normal Maps: A Generalized Solution for Depth Image, Point Cloud, and Mesh

Head pose estimation plays a crucial role in various applications, including human–machine interaction, autonomous driving systems, and 3D reconstruction. Current methods address the problem primarily from a 2D perspective, which limits the efficient utilization of 3D information. Herein, a novel approach, called pose orientation‐aware network (POANet), which leverages normal maps for orientation information embedding, providing abundant and robust head pose information, is introduced. POANet incorporates the axial signal perception module and the rotation matrix perception module, these lightweight modules make the approach achieve state‐of‐the‐art (SOTA) performance with few computational costs. This method can directly analyze various topological 3D data without extensive preprocessing. For depth images, POANet outperforms existing methods on the Biwi Kinect head pose dataset, reducing the mean absolute error (MAE) by ≈30% compared to the SOTA methods. POANet is the first method to perform rigid head registration in a landmark‐free manner. It also incorporates few‐shot learning capabilities and achieves an MAE of about on the Headspace dataset. These features make POANet a superior alternative to traditional generalized Procrustes analysis for mesh data processing, offering enhanced convenience for human phenotype studies.

Polyp segmentation network based on lightweight model and reverse attention mechanisms

AbstractColorectal cancer is a common gastrointestinal malignancy. Early screening and segmentation of colorectal polyps are of great clinical significance. Colonoscopy is the most effective method to detect polyps, but some polyps may be missed in the detection process. On this basis, the use of computer‐aided diagnosis technology is particularly important for colorectal polyp segmentation. To improve the detection rate of intestinal polyps under colonoscopy, a polyp segmentation network (MobileRaNet) based on a lightweight model and reverse attention (RA) mechanism was proposed to accurately segment polyps in colonoscopy images. The coordinated attention module is used to improve MobileNetV3 and make it the backbone network (CaNet). Second, a part of the output of the high‐level feature from the backbone network is passed into the parallel axial receptive field module (PA_RFB) to extract the global dependency representation without losing the details. Third, a global map is generated based on this combined feature as the initial boot area of the subsequent components. Finally, the RA module is used to mine the target region and boundary clues to improve the segmentation accuracy. To verify the effectiveness and lightweight performance of the algorithm, five challenging datasets, including CVC‐ColonDB, CVC‐300, and Kvasir, are used in this paper. In six indexes, including MeanDice, MeanIoU, and MAE, compared with seven typical models such as PraNet and TransUnet, accuracy, FLOPs, parameters, and FPS were compared. The experimental results show that the MobileRaNet proposed in this paper has improved the performance of the five datasets to varying degrees, especially the MeanDice and MeanIOU indexes of the Kvasir dataset reach 91.2% and 85.6%, which are, respectively, increased by 1.4% and 1.6% compared with PraNet. Compared with PraNet, FLOPs and parameters decreased by 83.3% and 76.7%, respectively.

MUPT-Net: Multi-scale U-shape pyramid transformer network for Infrared Small Target Detection

Infrared Small Target Detection (IRSTD) aims to detect small and dim targets in complex backgrounds. However, the low signal-to-noise ratio and reduced contrast in the infrared domain make it challenging to extract these targets, as the cluttered background can easily overpower them. Existing Convolutional Neural Networks (CNN)-based methods for IRSTD often suffer from information loss due to inadequate utilization of acquired information after downsampling operations. This limits their ability to accurately extract shape information related to infrared small targets. To address this challenge, we propose a Multi-scale U-shape Pyramid Transformer Network (MUPT-Net). Our network incorporates the U-shape Interaction Module (UIM) and the Multi-scale ViT Module (MSVM) to perform feature extraction. By fully leveraging and integrating the information obtained after each downsampling operation, our approach enables precise extraction of shape information for infrared small targets. Additionally, we introduce the Axial Compression Attention module (ACA), which focuses on capturing the interplay of positional information within the feature map to facilitate accurate detection of small targets. Through iteratively fusing and augmenting multi-scale features, our MUPT-Net effectively assimilates and harnesses contextual information of small targets. Experimental results on the SIRST v1, SIRST v2 and NUDT-SIRST datasets demonstrate the superiority of our approach compared to representative state-of-the-art (SOTA) IRSTD methods.

Attention Connect Network for Liver Tumor Segmentation from CT and MRI Images.

Introduction: Currently, the incidence of liver cancer is on the rise annually. Precise identification of liver tumors is crucial for clinicians to strategize the treatment and combat liver cancer. Thus far, liver tumor contours have been derived through labor-intensive and subjective manual labeling. Computers have gained widespread application in the realm of liver tumor segmentation. Nonetheless, liver tumor segmentation remains a formidable challenge owing to the diverse range of volumes, shapes, and image intensities encountered. Methods: In this article, we introduce an innovative solution called the attention connect network (AC-Net) designed for automated liver tumor segmentation. Building upon the U-shaped network architecture, our approach incorporates 2 critical attention modules: the axial attention module (AAM) and the vision transformer module (VTM), which replace conventional skip-connections to seamlessly integrate spatial features. The AAM facilitates feature fusion by computing axial attention across feature maps, while the VTM operates on the lowest resolution feature maps, employing multihead self-attention, and reshaping the output into a feature map for subsequent concatenation. Furthermore, we employ a specialized loss function tailored to our approach. Our methodology begins with pretraining AC-Net using the LiTS2017 dataset and subsequently fine-tunes it using computed tomography (CT) and magnetic resonance imaging (MRI) data sourced from Hubei Cancer Hospital. Results: The performance metrics for AC-Net on CT data are as follows: dice similarity coefficient (DSC) of 0.90, Jaccard coefficient (JC) of 0.82, recall of 0.92, average symmetric surface distance (ASSD) of 4.59, Hausdorff distance (HD) of 11.96, and precision of 0.89. For AC-Net on MRI data, the metrics are DSC of 0.80, JC of 0.70, recall of 0.82, ASSD of 7.58, HD of 30.26, and precision of 0.84. Conclusion: The comparative experiments highlight that AC-Net exhibits exceptional tumor recognition accuracy when tested on the Hubei Cancer Hospital dataset, demonstrating highly competitive performance for practical clinical applications. Furthermore, the ablation experiments provide conclusive evidence of the efficacy of each module proposed in this article. For those interested, the code for this research article can be accessed at the following GitHub repository: https://github.com/killian-zero/py_tumor-segmentation.git.

CGS‐Net: A classification‐guided framework for automated infection segmentation of COVID‐19 from CT images

AbstractAutomated segmentation of lung lesions in CT images of COVID‐19 based on deep learning holds great potential for comprehending the advancement of the disease and establishing suitable treatment approaches. However, the complex background, indistinct boundaries, varying sizes and distributions of infected regions, and high similarity to other lung diseases pose substantial challenges. To address these issues, we propose a joint deep learning‐based framework, named Classification‐Guided Segmentation Network (CGS‐Net), for COVID‐19 segmentation. The framework comprises a classification and a segmentation sub‐network, which are trained sequentially. Initially, the classification sub‐network learns the feature information of COVID‐19 and other pneumonia classifications. Subsequently, the trained classification sub‐network assists in the training of the segmentation sub‐network. In addition, several key modules are employed to construct the network, including the Multi‐scale Feature Mapping Module (MSFM), the Context Information Module (Context), and the Axial Attention Fusion Module (AAFM). The MSFM employs dilated convolutions to extract multi‐scale features with an emphasis on spatial and channel information. The Context combines strip pooling and average pooling to aggregate contextual and global information. Finally, the AAFM is employed to incorporate contextual information through an axial attention mechanism. Extensive experimental results demonstrate that our proposed method outperforms competing approaches in the segmentation task.

Geometric and Tooth Contact Analysis of the Areas and Perimeters, on the Archimedean Worm Wheels depending on the Modification of the Axial Module

Geometric and Tooth Contact Analysis of the Areas and Perimeters, on the Archimedean Worm Wheels depending on the Modification of the Axial Module

MAFormer: A New Method for Radar Reflectivity Reconstructing Using Satellite Data

Radar reflectivity plays a crucial role in detecting heavy rainfall and is an important tool for meteorological analysis. However, the coverage of a single radar is limited, leading to the use of satellite data as a complementary source. Consequently, how to bridge the gap between radar and satellite data has become a growing research focus. In this paper, we present MAFormer, a novel model for reconstructing radar reflectivity using satellite data within the Transformer framework. MAFormer consists of two modules: the Axial Local Attention Module and the Mixup Global Attention Module, which extract both local saliency and global similarity. Quantitative and qualitative experiments demonstrate the effectiveness of our proposed method. Specifically, the MAFormer model exhibits notable advancements when compared to state-of-the-art deep learning techniques. It demonstrates an improvement ranging from 0.01 to 0.05 in terms of the Heidke skill score, indicating its superior performance. Additionally, MAFormer effectively mitigates false alarm rates by approximately 0.016 to 0.04, which further highlights its enhanced accuracy and reliability.

GMRE-iUnet: Isomorphic Unet fusion model for PET and CT lung tumor images

Lung tumor PET and CT image fusion is a key technology in clinical diagnosis. However, the existing fusion methods are difficult to obtain fused images with high contrast, prominent morphological features, and accurate spatial localization. In this paper, an isomorphic Unet fusion model (GMRE-iUnet) for lung tumor PET and CT images is proposed to address the above problems. The main idea of this network is as following: Firstly, this paper constructs an isomorphic Unet fusion network, which contains two independent multiscale dual encoders Unet, it can capture the features of the lesion region, spatial localization, and enrich the morphological information. Secondly, a Hybrid CNN-Transformer feature extraction module (HCTrans) is constructed to effectively integrate local lesion features and global contextual information. In addition, the residual axial attention feature compensation module (RAAFC) is embedded into the Unet to capture fine-grained information as compensation features, which makes the model focus on local connections in neighboring pixels. Thirdly, a hybrid attentional feature fusion module (HAFF) is designed for multiscale feature information fusion, it aggregates edge information and detail representations using local entropy and Gaussian filtering. Finally, the experiment results on the multimodal lung tumor medical image dataset show that the model in this paper can achieve excellent fusion performance compared with other eight fusion models. In CT mediastinal window images and PET images comparison experiment, AG, EI, QAB/F, SF, SD, and IE indexes are improved by 16.19%, 26%, 3.81%, 1.65%, 3.91% and 8.01%, respectively. GMRE-iUnet can highlight the information and morphological features of the lesion areas and provide practical help for the aided diagnosis of lung tumors.

Axial compressive performance of ultra-high performance concrete-filled steel tube stub columns at different concrete age

To investigate the axial compressive performance of ultra-high-performance concrete-filled steel tube (UHPCFST) stub columns at different age, the axial compressive tests of UHPC specimens and UHPCFST short columns cast in the same batch at different concrete ages were carried out. The failure mode, stress–strain curve and the key indicators such as axial strength, elastic module and peak strains of UHPC at different ages were analysed. Meanwhile, the failure patterns, load–strain curves and the evolution of ultimate bearing capacities and stiffness of UHPCFST short columns at various ages were analysed. In addition, the constitutive relationship of mature normal concrete in the Chinese standard GB50010 was extended to numerically model UHPC columns under axial compression at different ages. Meanwhile, based on the axial compressive bearing capacity of UHPCFST short columns at mature age (28 days), a formula for calculating the load bearing capacity of UHPCFST short columns at different ages was proposed by introducing the age hooping correction coefficient, and verified using both test and simulation results. This study will provide a reference for the safe construction of ultra-high performance concrete-filled steel tubes.

Faster Region‐based Convolutional Neural Networks with You Only Look Once multi‐stage caries lesion from oral panoramic X‐ray images

AbstractNeural networks and artificial intelligence find more applications in dentistry. It treats dental caries, the most prevalent type of dental illness worldwide. Even though dental caries can be prevented and treated, they typically cause dental discomfort and tooth loss. For dental caries to be treated quickly and effectively, comprehensive detection may be needed, a combination of techniques that include eye inspection, probing, using a dental probe, and using a hand‐held mirror, and the individual application of each of these techniques, can quickly identify large caries cavities. Long‐established caries detection techniques help to locate only partially hidden but still accessible holes. Deep learning (DL) techniques have produced remarkable diagnostic results in radiology. This study aimed to classify various radiographic extensions on panoramic films using DL techniques, identify caries lesions using these techniques, and compare the results to those of dentists with extensive training. Faster region‐based convolutional neural networks (R‐CNN) is a newly discovered field of medical research that is rapidly expanding and has produced outstanding results in diagnosing and prognosis of pathology and radiology conditions. In this study, dental cavities were detected and analysed using periapical radiographs to evaluate the accuracy of the Faster R‐CNN algorithm. Because these three caries were derived from the oral panoramic images, we designed You Only Look Once Version 3 (YOLOv3) as a U‐shaped network with a large‐scale axial attention module. We also compare the effectiveness of YOLOv3's segmentation to that of other industrial standards. Experiments show that our proposed method, Fast R‐CNN–YOLOv3, achieves higher accuracy in segmenting the three distinct caries level. The proposed model (R‐CNN–YOLOv3) achieved an effective result with a precision of 97.183%.

ASA-DRNet: An Improved Deeplabv3+ Framework for SAR Image Segmentation

Pollution caused by oil spills does irreversible harm to marine biosystems. To find maritime oil spills, Synthetic Aperture Radar (SAR) has emerged as a crucial mean. How to accurately distinguish oil spill areas from other types of areas is a committed step in detecting oil spills. Owing to its capacity to extract multiscale features and its distinctive decoder, the Deeplabv3+ framework has been developed into an excellent deep learning model in field of picture segmentation. However, in some SAR pictures, there is a lack of clarity in the segmentation of oil film edges and incorrect segmentation of small areas. In order to solve these problems, an improved network, named ASA-DRNet, has been proposed. Firstly, a new structure which combines an axial self-attention module with ResNet-18 is proposed as the backbone of DeepLabv3+ encoder. Secondly, a atrous spatial pyramid pooling (ASPP) module is optimized to improve the network’s capacity of extracting multiscale features and to increase the speed of model calculation and finally merging low-level features of different resolutions to enhance the competence of network to extract edge information. The experiments show that ASA-DRNet obtains the better results compared to other neural network models.

CCT: Lightweight compact convolutional transformer for lung disease CT image classification.

Computed tomography (CT) imaging results are an important criterion for the diagnosis of lung disease. CT images can clearly show the characteristics of lung lesions. Early and accurate detection of lung diseases helps clinicians to improve patient care effectively. Therefore, in this study, we used a lightweight compact convolutional transformer (CCT) to build a prediction model for lung disease classification using chest CT images. We added a position offset term and changed the attention mechanism of the transformer encoder to an axial attention mechanism module. As a result, the classification performance of the model was improved in terms of height and width. We show that the model effectively classifies COVID-19, community pneumonia, and normal conditions on the CC-CCII dataset. The proposed model outperforms other comparable models in the test set, achieving an accuracy of 98.5% and a sensitivity of 98.6%. The results show that our method achieves a larger field of perception on CT images, which positively affects the classification of CT images. Thus, the method can provide adequate assistance to clinicians.

Estimating the initial stage in the process of radial-reverse extrusion using a triangular kinematic module

Those parts of solid or hollow blanks whose shape is complex should be produced by means of combined radial-longitudinal extrusion. However, the use of combined extrusion processes with several degrees of flow freedom requires a preliminary assessment of shape formation, which is true, taking into consideration the peculiarities of evolution of strain sites at different stages of deformation. When deforming high blanks, the presence of an intermediate rigid zone can be observed, separating two autonomous strain sites. When constructing an estimation scheme of the initial stage of the process of combined radial-backward extrusion of hollow parts with a flange, the presence of an intermediate rigid zone is taken into consideration. The need to improve the devised estimation scheme is caused by significant deviations in the projected growths of a part from its experimentally derived dimensions. As an alternative to the axial rectangular kinematic module of the lower deformation site, the use of an axial triangular module has been proposed, whose effectiveness is demonstrated in simulating the process of radial-longitudinal extrusion with expansion. The rationality of the proposed replacement was revealed, both for forecasting the forced mode of the deformation process and for the gradual part’s shape formation. This has made it possible to reduce the projected estimates to 10 % in terms of the increase in the size of a part based on a comparative analysis with experimentally derived data. It is recommended to use the devised scheme for modeling the initial stage of the process for relatively high blanks at H0/h1>4…6; the limitation is the degeneration of the intermediate rigid zone. This will contribute to compiling recommendations for expanding the possibilities of using combined radial-backward extrusion of hollow parts with a flange during production

Earf-YOLO: An Efficient Attention Receptive Field Model for Recognizing Symbols of Zhuang Minority Patterns

As for recognizing Zhuang minority pattern symbols, current recognition models often cause high computational overhead and low accuracy since Zhuang minority pattern symbols have large feature vectors and some complex features. In this paper, we present the efficient attention receptive field you only look once (Earf-YOLO), a new scheme to address those problems. Firstly, a global-local-transformer (GLocalT) structure is proposed, through which other control systems are introduced into the axial self-attention module, and global-local training strategies are also designed. The structure can use other control systems to compensate for the lost feature information along the height, width, and channel axes. The global-local training strategy can encode long-term dependencies between features and reduce local information loss, fully illustrating that the structure has high feature expression ability. Besides, strength receptive field block (SRFB) is suggested to use the dilated convolution to control the receptive field’s eccentricity and enrich the feature information of the receptive field during its training. With more branches, it can better extract multiscale features, enrich the feature space of the convolution block, and reparametrize multibranch during prediction to fuse them into the main branch, all of which contribute to the improvement of the model performance. Finally, some advanced training techniques are adopted to enhance the detection effect further. In the end, comparative experiments are conducted on the datasets of Zhuang pattern symbols and PASCAL VOC, whose results indicate that the AP and FPS of the suggested model reach their highest values, manifesting its high efficiency.

CariesNet: a deep learning approach for segmentation of multi-stage caries lesion from oral panoramic X-ray image.

Dental caries has been a common health issue throughout the world, which can even lead to dental pulp and root apical inflammation eventually. Timely and effective treatment of dental caries is vital for patients to reduce pain. Traditional caries disease diagnosis methods like naked-eye detection and panoramic radiograph examinations rely on experienced doctors, which may cause misdiagnosis and high time-consuming. To this end, we propose a novel deep learning architecture called CariesNet to delineate different caries degrees from panoramic radiographs. We firstly collect a high-quality panoramic radiograph dataset with 3127 well-delineated caries lesions, including shallow caries, moderate caries, and deep caries. Then we construct CariesNet as a U-shape network with the additional full-scale axial attention module to segment these three caries types from the oral panoramic images. Moreover, we test the segmentation performance between CariesNet and other baseline methods. Experiments show that our method can achieve a mean 93.64% Dice coefficient and 93.61% accuracy in the segmentation of three different levels of caries.

Hyperelastic Membrane Actuators: Analysis of Toroidal and Helical Multifunctional Configurations.

Technologies that provide mechanical assistance are required in the medical field, such as implants that regenerate tissue through elongation and stimulation. One of the challenges is to develop actuators that combine the benefits of high axial extension at low pressures, modularity, multifunction, and load-bearing capabilities into one design while maintaining their shape and softness. Overcoming such a challenge will provide implants with enhanced capacity for mechanical assistance to induce tissue regeneration. We introduce two novel actuators (M2H) built of stacked Hyperelastic Ballooning Membrane Actuators (HBMAs) that can be realized using helical and toroidal configurations. By restraining the HBMA expansion deterministically using a semisoft exoskeleton, the actuators are endowed with axial extension and radial expansion capabilities. These actuators are thus built of modules that can be configured to different therapeutical needs and multifunctionality, to provide anatomically congruent stimulation. We present the design, fabrication, testing, and numerical and experimental validation of the M2H-HBMAs. They can axially extend up to 41% and 32% in their helical and toroidal configurations at input pressures as low as 26 and 24 kPa, respectively. If the axial extension module is used separately, its extension capacity reaches >170%. The M2H-HBMAs can perform independent and simultaneous expansion and extension motions with negligible intraluminal deformation as well as stand at least 1 kg of axial force without collapsing. The M2H-HBMAs overcome the limitations of hyperexpanding machines that show low resistance to load. We envisage M2H-HBMAs as promising tools to perform tissue regeneration procedures.

ATMConvGRU for Weather Forecasting

Weather forecasting, which is challenging due to the complex atmospheric correlation, focuses on providing explicit meteorological estimations as accurate as possible. Recently, techniques based upon convolutional-recurrent networks have shown dramatic performance in domains including radar echo prediction and precipitation forecasting, indicating that deep learning models have great potential in this area. However, existing methods concentrate on adding extra paralleled memory cells to the inner recurrent unit, where the information is mutually independent. To extract spatial–temporal features with stronger correlation, this letter introduced an axial attention memory module with quasi state-in-state cascaded manner. Benefitting from this unit, the spatial–temporal features can be aggregated and embedded into standard ConvGRU formulating Axial aTtention Memory Cascaded ConvGRU (ATMConvGRU) for sequential weather prediction. Experimental results compared with many state-of-the-art methods on four types of weather forecasting related datasets, including temperature, relative humidity, wind, and radar echo, demonstrate the effectiveness and superiority of the proposed ATMConvGRU.