MMDB: Multimodal dual-branch model for multi-functional bioactive peptide prediction.

Bioactive peptides can hinder oxidative processes and microbial spoilage in foodstuffs and play important roles in treating diverse diseases and disorders. While most of the methods focus on single-functional bioactive peptides and have obtained promising prediction performance, it is still a significant challenge to accurately detect complex and diverse functions simultaneously with the quick increase of multi-functional bioactive peptides. In contrast to previous research on multi-functional bioactive peptide prediction based solely on sequence, we propose a novel multimodal dual-branch (MMDB) lightweight deep learning model that designs two different branches to effectively capture the complementary information of peptide sequence and structural properties. Specifically, a multi-scale dilated convolution with Bi-LSTM branch is presented to effectively model the different scales sequence properties of peptides while a multi-layer convolution branch is proposed to capture structural information. To the best of our knowledge, this is the first effective extraction of peptide sequence features using multi-scale dilated convolution without parameter increase. Multimodal features from both branches are integrated via a fully connected layer for multi-label classification. Compared to state-of-the-art methods, our MMDB model exhibits competitive results across metrics, with a 9.1% Coverage increase and 5.3% and 3.5% improvements in Precision and Accuracy, respectively.

ASCEND-UNet: An Improved UNet Configuration Optimized for Rural Settlements Mapping.

Different types of rural settlement agglomerations have been formed and mixed in space during the rural revitalization strategy implementation in China. Discriminating them from remote sensing images is of great significance for rural land planning and living environment improvement. Currently, there is a lack of automatic methods for obtaining information on rural settlement differentiation. In this paper, an improved encoder-decoder network structure, ASCEND-UNet, was designed based on the original UNet. It was implemented to segment and classify dispersed and clustered rural settlement buildings from high-resolution satellite images. The ASCEND-UNet model incorporated three components: firstly, the atrous spatial pyramid pooling (ASPP) multi-scale feature fusion module was added into the encoder, then the spatial and channel squeeze and excitation (scSE) block was embedded at the skip connection; thirdly, the hybrid dilated convolution (HDC) block was utilized in the decoder. In our proposed framework, the ASPP and HDC were used as multiple dilated convolution blocks to expand the receptive field by introducing a series of dilated rate convolutions. The scSE is an attention mechanism block focusing on features both in the spatial and channel dimension. A series of model comparisons and accuracy assessments with the original UNet, PSPNet, DeepLabV3+, and SegNet verified the effectiveness of our proposed model. Compared with the original UNet model, ASCEND-UNet achieved improvements of 4.67%, 2.80%, 3.73%, and 6.28% in precision, recall, F1-score and MIoU, respectively. The contributions of HDC, ASPP, and scSE modules were discussed in ablation experiments. Our proposed model obtained more accurate and stable results by integrating multiple dilated convolution blocks with an attention mechanism. This novel model enriches the automatic methods for semantic segmentation of different rural settlements from remote sensing images.

Research into the Applications of a Multi-Scale Feature Fusion Model in the Recognition of Abnormal Human Behavior.

Due to the increasing severity of aging populations in modern society, the accurate and timely identification of, and responses to, sudden abnormal behaviors of the elderly have become an urgent and important issue. In the current research on computer vision-based abnormal behavior recognition, most algorithms have shown poor generalization and recognition abilities in practical applications, as well as issues with recognizing single actions. To address these problems, an MSCS-DenseNet-LSTM model based on a multi-scale attention mechanism is proposed. This model integrates the MSCS (Multi-Scale Convolutional Structure) module into the initial convolutional layer of the DenseNet model to form a multi-scale convolution structure. It introduces the improved Inception X module into the Dense Block to form an Inception Dense structure, and gradually performs feature fusion through each Dense Block module. The CBAM attention mechanism module is added to the dual-layer LSTM to enhance the model's generalization ability while ensuring the accurate recognition of abnormal actions. Furthermore, to address the issue of single-action abnormal behavior datasets, the RGB image dataset RIDS (RGB image dataset) and the contour image dataset CIDS (contour image dataset) containing various abnormal behaviors were constructed. The experimental results validate that the proposed MSCS-DenseNet-LSTM model achieved an accuracy, sensitivity, and specificity of 98.80%, 98.75%, and 98.82% on the two datasets, and 98.30%, 98.28%, and 98.38%, respectively.

Multi-Branch Network for Color Image Denoising Using Dilated Convolution and Attention Mechanisms.

Image denoising is regarded as an ill-posed problem in computer vision tasks that removes additive noise from imaging sensors. Recently, several convolution neural network-based image-denoising methods have achieved remarkable advances. However, it is difficult for a simple denoising network to recover aesthetically pleasing images owing to the complexity of image content. Therefore, this study proposes a multi-branch network to improve the performance of the denoising method. First, the proposed network is designed based on a conventional autoencoder to learn multi-level contextual features from input images. Subsequently, we integrate two modules into the network, including the Pyramid Context Module (PCM) and the Residual Bottleneck Attention Module (RBAM), to extract salient information for the training process. More specifically, PCM is applied at the beginning of the network to enlarge the receptive field and successfully address the loss of global information using dilated convolution. Meanwhile, RBAM is inserted into the middle of the encoder and decoder to eliminate degraded features and reduce undesired artifacts. Finally, extensive experimental results prove the superiority of the proposed method over state-of-the-art deep-learning methods in terms of objective and subjective performances.

Surface Defect Detection of Aluminum Profiles Based on Multiscale and Self-Attention Mechanisms.

To address the various challenges in aluminum surface defect detection, such as multiscale intricacies, sensitivity to lighting variations, occlusion, and noise, this study proposes the AluDef-ClassNet model. Firstly, a Gaussian difference pyramid is utilized to capture multiscale image features. Secondly, a self-attention mechanism is introduced to enhance feature representation. Additionally, an improved residual network structure incorporating dilated convolutions is adopted to increase the receptive field, thereby enhancing the network's ability to learn from extensive information. A small-scale dataset of high-quality aluminum surface defect images is acquired using a CCD camera. To better tackle the challenges in surface defect detection, advanced deep learning techniques and data augmentation strategies are employed. To address the difficulty of data labeling, a transfer learning approach based on fine-tuning is utilized, leveraging prior knowledge to enhance the efficiency and accuracy of model training. In dataset testing, our model achieved a classification accuracy of 97.6%, demonstrating significant advantages over other classification models.

Plant leaf infected spot segmentation using robust encoder-decoder cascaded deep learning model.

Leaf infection detection and diagnosis at an earlier stage can improve agricultural output and reduce monetary costs. An inaccurate segmentation may degrade the accuracy of disease classification due to some different and complex leaf diseases. Also, the disease's adhesion and dimension can overlap, causing partial under-segmentation. Therefore, a novel robust Deep Encoder-Decoder Cascaded Network (DEDCNet) model is proposed in this manuscript for leaf image segmentation that precisely segments the diseased leaf spots and differentiates similar diseases. This model is comprised of an Infected Spot Recognition Network and an Infected Spot Segmentation Network. Initially, ISRN is designed by integrating cascaded CNN with a Feature Pyramid Pooling layer to identify the infected leaf spot and avoid an impact of background details. After that, the ISSN developed using an encoder-decoder network, which uses a multi-scale dilated convolution kernel to precisely segment the infected leaf spot. Moreover, the resultant leaf segments are provided to the pre-learned CNN models to learn texture features followed by the SVM algorithm to categorize leaf disease classes. The ODEDCNet delivers exceptional performance on both the Betel Leaf Image and PlantVillage datasets. On the Betel Leaf Image dataset, it achieves an accuracy of 94.89%, with high precision (94.35%), recall (94.77%), and F-score (94.56%), while maintaining low under-segmentation (6.2%) and over-segmentation rates (2.8%). It also achieves a remarkable Dice coefficient of 0.9822, all in just 0.10 seconds. On the PlantVillage dataset, the ODEDCNet outperforms other existing models with an accuracy of 96.5%, demonstrating high precision (96.61%), recall (96.5%), and F-score (96.56%). It excels in reducing under-segmentation to just 3.12% and over-segmentation to 2.56%. Furthermore, it achieves a Dice coefficient of 0.9834 in a mere 0.09 seconds. It evident for the greater efficiency on both segmentation and categorization of leaf diseases contrasted with the existing models.

Contrast-enhanced MRI synthesis using dense-dilated residual convolutions based 3D network toward elimination of gadolinium in neuro-oncology.

Recent studies have raised broad safety and health concerns about using of gadolinium contrast agents during magnetic resonance imaging (MRI) to enhance identification of active tumors. In this paper, we developed a deep learning-based method for three-dimensional (3D) contrast-enhanced T1-weighted (T1) image synthesis from contrast-free image(s). The MR images of 1251 patients with glioma from the RSNA-ASNR-MICCAI BraTS Challenge 2021 dataset were used in this study. A 3D dense-dilated residual U-Net (DD-Res U-Net) was developed for contrast-enhanced T1 image synthesis from contrast-free image(s). The model was trained on a randomly split training set (n = 800) using a customized loss function and validated on a validation set (n = 200) to improve its generalizability. The generated images were quantitatively assessed against the ground-truth on a test set (n = 251) using the mean absolute error (MAE), mean-squared error (MSE), peak signal-to-noise ratio (PSNR), structural similarity (SSIM), normalized mutual information (NMI), and Hausdorff distance (HDD) metrics. We also performed a qualitative visual similarity assessment between the synthetic and ground-truth images. The effectiveness of the proposed model was compared with a 3D U-Net baseline model and existing deep learning-based methods in the literature. Our proposed DD-Res U-Net model achieved promising performance for contrast-enhanced T1 synthesis in both quantitative metrics and perceptual evaluation on the test set (n = 251). Analysis of results on the whole brain region showed a PSNR (in dB) of 29.882 ± 5.924, a SSIM of 0.901 ± 0.071, a MAE of 0.018 ± 0.013, a MSE of 0.002 ± 0.002, a HDD of 2.329 ± 9.623, and a NMI of 1.352 ± 0.091 when using only T1 as input; and a PSNR (in dB) of 30.284 ± 4.934, a SSIM of 0.915 ± 0.063, a MAE of 0.017 ± 0.013, a MSE of 0.001 ± 0.002, a HDD of 1.323 ± 3.551, and a NMI of 1.364 ± 0.089 when combining T1 with other MRI sequences. Compared to the U-Net baseline model, our model revealed superior performance. Our model demonstrated excellent capability in generating synthetic contrast-enhanced T1 images from contrast-free MR image(s) of the whole brain region when using multiple contrast-free images as input. Without incorporating tumor mask information during network training, its performance was inferior in the tumor regions compared to the whole brain which requires further improvements to replace the gadolinium administration in neuro-oncology.

E2LNet: An Efficient and Effective Lightweight Network for Panoramic Depth Estimation.

Monocular panoramic depth estimation has various applications in robotics and autonomous driving due to its ability to perceive the entire field of view. However, panoramic depth estimation faces two significant challenges: global context capturing and distortion awareness. In this paper, we propose a new framework for panoramic depth estimation that can simultaneously address panoramic distortion and extract global context information, thereby improving the performance of panoramic depth estimation. Specifically, we introduce an attention mechanism into the multi-scale dilated convolution and adaptively adjust the receptive field size between different spatial positions, designing the adaptive attention dilated convolution module, which effectively perceives distortion. At the same time, we design the global scene understanding module to integrate global context information into the feature maps generated using the feature extractor. Finally, we trained and evaluated our model on three benchmark datasets which contains the virtual and real-world RGB-D panorama datasets. The experimental results show that the proposed method achieves competitive performance, comparable to existing techniques in both quantitative and qualitative evaluations. Furthermore, our method has fewer parameters and more flexibility, making it a scalable solution in mobile AR.

IRDC-Net: An Inception Network with a Residual Module and Dilated Convolution for Sign Language Recognition Based on Surface Electromyography.

Deaf and hearing-impaired people always face communication barriers. Non-invasive surface electromyography (sEMG) sensor-based sign language recognition (SLR) technology can help them to better integrate into social life. Since the traditional tandem convolutional neural network (CNN) structure used in most CNN-based studies inadequately captures the features of the input data, we propose a novel inception architecture with a residual module and dilated convolution (IRDC-net) to enlarge the receptive fields and enrich the feature maps, applying it to SLR tasks for the first time. This work first transformed the time domain signal into a time-frequency domain using discrete Fourier transformation. Second, an IRDC-net was constructed to recognize ten Chinese sign language signs. Third, the tandem CNN networks VGG-net and ResNet-18 were compared with our proposed parallel structure network, IRDC-net. Finally, the public dataset Ninapro DB1 was utilized to verify the generalization performance of the IRDC-net. The results showed that after transforming the time domain sEMG signal into the time-frequency domain, the classification accuracy (acc) increased from 84.29% to 91.70% when using the IRDC-net on our sign language dataset. Furthermore, for the time-frequency information of the public dataset Ninapro DB1, the classification accuracy reached 89.82%; this value is higher than that achieved in other recent studies. As such, our findings contribute to research into SLR tasks and to improving deaf and hearing-impaired people's daily lives.

MAB-DrNet: Bearing Fault Diagnosis Method Based on an Improved Dilated Convolutional Neural Network.

Rolling bearing fault diagnosis is of great significance to the safe and reliable operation of manufacturing equipment. In the actual complex environment, the collected bearing signals usually contain a large amount of noises from the resonances of the environment and other components, resulting in the nonlinear characteristics of the collected data. Existing deep-learning-based solutions for bearing fault diagnosis perform poorly in classification performance under noises. To address the above problems, this paper proposes an improved dilated-convolutional-neural network-based bearing fault diagnosis method in noisy environments named MAB-DrNet. First, a basic model called the dilated residual network (DrNet) was designed based on the residual block to enlarge the model's perceptual field to better capture the features from bearing fault signals. Then, a max-average block (MAB) module was designed to improve the feature extraction capability of the model. In addition, the global residual block (GRB) module was introduced into MAB-DrNet to further improve the performance of the proposed model, enabling the model to better handle the global information of the input data and improve the classification accuracy of the model in noisy environments. Finally, the proposed method was tested on the CWRU dataset, and the results showed that the proposed method had good noise immunity; the accuracy was 95.57% when adding Gaussian white noises with a signal-to-noise ratio of -6 dB. The proposed method was also compared with existing advanced methods to further prove its high accuracy.

Intelligent Bearing Fault Diagnosis Based on Feature Fusion of One-Dimensional Dilated CNN and Multi-Domain Signal Processing.

Finding relevant features that can represent different types of faults under a noisy environment is the key to practical applications of intelligent fault diagnosis. However, high classification accuracy cannot be achieved with only a few simple empirical features, and advanced feature engineering and modelling necessitate extensive specialised knowledge, resulting in restricted widespread use. This paper has proposed a novel and efficient fusion method, named MD-1d-DCNN, that combines statistical features from multiple domains and adaptive features retrieved using a one-dimensional dilated convolutional neural network. Moreover, signal processing techniques are utilised to uncover statistical features and realise the general fault information. To offset the negative influence of noise in signals and achieve high accuracy of fault diagnosis in noisy settings, 1d-DCNN is adopted to extract more dispersed and intrinsic fault-associated features, while also preventing the model from overfitting. In the end, fault classification based on fusion features is accomplished by the usage of fully connected layers. Two bearing datasets containing varying amounts of noise are used to verify the effectiveness and robustness of the suggested approach. The experimental results demonstrate MD-1d-DCNN's superior anti-noise capability. When compared to other benchmark models, the proposed method performs better at all noise levels.

Faster SCDNet: Real-Time Semantic Segmentation Network with Split Connection and Flexible Dilated Convolution.

Recently, semantic segmentation has been widely applied in various realistic scenarios. Many semantic segmentation backbone networks use various forms of dense connection to improve the efficiency of gradient propagation in the network. They achieve excellent segmentation accuracy but lack inference speed. Therefore, we propose a backbone network SCDNet with a dual path structure and higher speed and accuracy. Firstly, we propose a split connection structure, which is a streamlined lightweight backbone with a parallel structure to increase inference speed. Secondly, we introduce a flexible dilated convolution using different dilation rates so that the network can have richer receptive fields to perceive objects. Then, we propose a three-level hierarchical module to effectively balance the feature maps with multiple resolutions. Finally, a refined flexible and lightweight decoder is utilized. Our work achieves a trade-off of accuracy and speed on the Cityscapes and Camvid datasets. Specifically, we obtain a 36% improvement in FPS and a 0.7% improvement in mIoU on the Cityscapes test set.

PEIPNet: Parametric Efficient Image-Inpainting Network with Depthwise and Pointwise Convolution.

Research on image-inpainting tasks has mainly focused on enhancing performance by augmenting various stages and modules. However, this trend does not consider the increase in the number of model parameters and operational memory, which increases the burden on computational resources. To solve this problem, we propose a Parametric Efficient Image InPainting Network (PEIPNet) for efficient and effective image-inpainting. Unlike other state-of-the-art methods, the proposed model has a one-stage inpainting framework in which depthwise and pointwise convolutions are adopted to reduce the number of parameters and computational cost. To generate semantically appealing results, we selected three unique components: spatially-adaptive denormalization (SPADE), dense dilated convolution module (DDCM), and efficient self-attention (ESA). SPADE was adopted to conditionally normalize activations according to the mask in order to distinguish between damaged and undamaged regions. The DDCM was employed at every scale to overcome the gradient-vanishing obstacle and gradually fill in the pixels by capturing global information along the feature maps. The ESA was utilized to obtain clues from unmasked areas by extracting long-range information. In terms of efficiency, our model has the lowest operational memory compared with other state-of-the-art methods. Both qualitative and quantitative experiments demonstrate the generalized inpainting of our method on three public datasets: Paris StreetView, CelebA, and Places2.

DDCNet: Deep Dilated Convolutional Neural Network for Dense Prediction.

Dense pixel matching problems such as optical flow and disparity estimation are among the most challenging tasks in computer vision. Recently, several deep learning methods designed for these problems have been successful. A sufficiently larger effective receptive field (ERF) and a higher resolution of spatial features within a network are essential for providing higher-resolution dense estimates. In this work, we present a systemic approach to design network architectures that can provide a larger receptive field while maintaining a higher spatial feature resolution. To achieve a larger ERF, we utilized dilated convolutional layers. By aggressively increasing dilation rates in the deeper layers, we were able to achieve a sufficiently larger ERF with a significantly fewer number of trainable parameters. We used optical flow estimation problem as the primary benchmark to illustrate our network design strategy. The benchmark results (Sintel, KITTI, and Middlebury) indicate that our compact networks can achieve comparable performance in the class of lightweight networks.

DFR-U-Net: Dual residual and feature fusion network for ulna and radius segmentation on dual-energy X-ray absorptiometry images.

BACKGROUND: Ulna and radius segmentation of dual-energy X-ray absorptiometry (DXA) images is essential for measuring bone mineral density (BMD). OBJECTIVE: To develop and test a novel deep learning network architecture for robust and efficient ulna and radius segmentation on DXA images. METHODS: This study used two datasets including 360 cases. The first dataset included 300 cases that were randomly divided into five groups for five-fold cross-validation. The second dataset including 60 cases was used for independent testing. A deep learning network architecture with dual residual dilated convolution module and feature fusion block based on residual U-Net (DFR-U-Net) to enhance segmentation accuracy of ulna and radius regions on DXA images was developed. The Dice similarity coefficient (DSC), Jaccard, and Hausdorff distance (HD) were used to evaluate the segmentation performance. A one-tailed paired t-test was used to assert the statistical significance of our method and the other deep learning-based methods (P < 0.05 indicates a statistical significance). RESULTS: The results demonstrated our method achieved the promising segmentation performance, with DSC of 98.56±0.40% and 98.86±0.25%, Jaccard of 97.14±0.75% and 97.73±0.48%, and HD of 6.41±11.67 pixels and 8.23±7.82 pixels for segmentation of ulna and radius, respectively. According to statistics data analysis results, our method yielded significantly higher performance than other deep learning-based methods. CONCLUSIONS: The proposed DFR-U-Net achieved higher segmentation performance for ulna and radius on DXA images than the previous work and other deep learning approaches. This methodology has potential to be applied to ulna and radius segmentation to help doctors measure BMD more accurately in the future.

Unsupervised Low-Light Image Enhancement Based on Generative Adversarial Network.

Low-light image enhancement aims to improve the perceptual quality of images captured under low-light conditions. This paper proposes a novel generative adversarial network to enhance low-light image quality. Firstly, it designs a generator consisting of residual modules with hybrid attention modules and parallel dilated convolution modules. The residual module is designed to prevent gradient explosion during training and to avoid feature information loss. The hybrid attention module is designed to make the network pay more attention to useful features. A parallel dilated convolution module is designed to increase the receptive field and capture multi-scale information. Additionally, a skip connection is utilized to fuse shallow features with deep features to extract more effective features. Secondly, a discriminator is designed to improve the discrimination ability. Finally, an improved loss function is proposed by incorporating pixel loss to effectively recover detailed information. The proposed method demonstrates superior performance in enhancing low-light images compared to seven other methods.

MTDCNet: A 3D multi-threading dilated convolutional network for brain tumor automatic segmentation.

Glioma is one of the most threatening tumors and the survival rate of the infected patient is low. The automatic segmentation of the tumors by reliable algorithms can reduce diagnosis time. In this paper, a novel 3D multi-threading dilated convolutional network (MTDC-Net) is proposed for the automatic brain tumor segmentation. First of all, a multi-threading dilated convolution (MTDC) strategy is introduced in the encoder part, so that the low dimensional structural features can be extracted and integrated better. At the same time, the pyramid matrix fusion (PMF) algorithm is used to integrate the characteristic structural information better. Secondly, in order to make the better use of context semantical information, this paper proposed a spatial pyramid convolution (SPC) operation. By using convolution with different kernel sizes, the model can aggregate more semantic information. Finally, the multi-threading adaptive pooling up-sampling (MTAU) strategy is used to increase the weight of semantic information, and improve the recognition ability of the model. And a pixel-based post-processing method is used to prevent the effects of error prediction. On the brain tumors segmentation challenge 2018 (BraTS2018) public validation dataset, the dice scores of MTDC-Net are 0.832, 0.892 and 0.809 for core, whole and enhanced of the tumor, respectively. On the BraTS2020 public validation dataset, the dice scores of MTDC-Net are 0.833, 0.896 and 0.797 for the core tumor, whole tumor and enhancing tumor, respectively. Mass numerical experiments show that MTDC-Net is a state-of-the-art network for automatic brain tumor segmentation.

ThelR547v1-An Asymmetric Dilated Convolutional Neural Network for Real-time Semantic Segmentation of Horticultural Crops.

Robust and automated image segmentation in high-throughput image-based plant phenotyping has received considerable attention in the last decade. The possibility of this approach has not been well studied due to the time-consuming manual segmentation and lack of appropriate datasets. Segmenting images of greenhouse and open-field grown crops from the background is a challenging task linked to various factors such as complex background (presence of humans, equipment, devices, and machinery for crop management practices), environmental conditions (humidity, cloudy/sunny, fog, rain), occlusion, low-contrast and variability in crops and pose over time. This paper presents a new ubiquitous deep learning architecture ThelR547v1 (Thermal RGB 547 layers version 1) that segmented each pixel as crop or crop canopy from the background (non-crop) in real time by abstracting multi-scale contextual information with reduced memory cost. By evaluating over 37,328 augmented images (aug1: thermal RGB and RGB), our method achieves mean IoU of 0.94 and 0.87 for leaves and background and mean Bf scores of 0.93 and 0.86, respectively. ThelR547v1 has a training accuracy of 96.27%, a training loss of 0.09, a validation accuracy of 96.15%, and a validation loss of 0.10. Qualitative analysis further shows that despite the low resolution of training data, ThelR547v1 successfully distinguishes leaf/canopy pixels from complex and noisy background pixels, enabling it to be used for real-time semantic segmentation of horticultural crops.

State of Health Prediction of Lithium-Ion Battery Based on Deep Dilated Convolution.

A battery's charging data include the timing information with respect to the charge. However, the existing State of Health (SOH) prediction methods rarely consider this information. This paper proposes a dilated convolution-based SOH prediction model to verify the influence of charging timing information on SOH prediction results. The model uses holes to fill in the standard convolutional kernel in order to expand the receptive field without adding parameters, thereby obtaining a wider range of charging timing information. Experimental data from six batteries of the same battery type were used to verify the model's effectiveness under different experimental conditions. The proposed method is able to accurately predict the battery SOH value in any range of voltage input through cross-validation, and the SDE (standard deviation of the error) is at least 0.28% lower than other methods. In addition, the influence of the position and length of the range of input voltage on the model's prediction ability is studied as well. The results of our analysis show that the proposed method is robust to different sampling positions and different sampling lengths of input data, which solves the problem of the original data being difficult to obtain due to the uncertainty of charging-discharging behaviour in actual operation.

RDCTrans U-Net: A Hybrid Variable Architecture for Liver CT Image Segmentation.

Segmenting medical images is a necessary prerequisite for disease diagnosis and treatment planning. Among various medical image segmentation tasks, U-Net-based variants have been widely used in liver tumor segmentation tasks. In view of the highly variable shape and size of tumors, in order to improve the accuracy of segmentation, this paper proposes a U-Net-based hybrid variable structure-RDCTrans U-Net for liver tumor segmentation in computed tomography (CT) examinations. We design a backbone network dominated by ResNeXt50 and supplemented by dilated convolution to increase the network depth, expand the perceptual field, and improve the efficiency of feature extraction without increasing the parameters. At the same time, Transformer is introduced in down-sampling to increase the network's overall perception and global understanding of the image and to improve the accuracy of liver tumor segmentation. The method proposed in this paper tests the segmentation performance of liver tumors on the LiTS (Liver Tumor Segmentation) dataset. It obtained 89.22% mIoU and 98.91% Acc, for liver and tumor segmentation. The proposed model also achieved 93.38% Dice and 89.87% Dice, respectively. Compared with the original U-Net and the U-Net model that introduces dense connection, attention mechanism, and Transformer, respectively, the method proposed in this paper achieves SOTA (state of art) results.