检索结果-内蒙古大学图书馆

EdGE: Edge distillation and gap elimination for heterogeneous networks in 3d medical image segmentation

KNOWLEdGE-BASEd SYSTEMS 2025年 314卷

作者： Yu, Xiangchun Wu, Tianqi Zhang, dingwen Zheng, Jian Wu, Jianqing Jiangxi Univ Sci & Technol Sch Informat Engn Jiangxi Prov Key Lab Multidimens Intelligent Perce Ganzhou 341000 Peoples R China

Compressing the cumbersome Vision Transformers (ViTs) or ConvNets into compact students can effectively facilitate the deployment of 3d medical image segmentation models on embedded devices. However, teacherstudent heterogeneity coupled with huge capacity discrepancies poses tremendous challenges in facilitating effective knowledge transfer. This paper introduces the Edge distillation and Gap Elimination (EdGE) framework for 3d medical image segmentation while effectively mitigating the knowledge gap across heterogeneous networks. We observe that the degradation of knowledge transfer between heterogeneous networks in 3d medical image segmentation may be attributed to a lack of edge-sensitive knowledge transfer. Initially, Edge Constraint Knowledge distillation (ECKd) is introduced to facilitate knowledge transfer of edge-sensitive logits and enhance the student's edge perception ability. Subsequently, we propose quantifying the knowledge gap as Hausdorff distance (Hd) in 3d abdominal CT images. Accordingly, segmentation Refinement Knowledge distillation (SRKd) is further introduced after the ECKd module to address knowledge transfer degradation in heterogeneous networks by minimizing the Hd. Lastly, Scale Adaptation Knowledge distillation (SAKd) is presented to enhance the student's performance in segmenting multi-organs with varying scales after projecting multiple stages of features into logits space. Extensive experiments are conducted on two 3d segmentation datasets, WORd and BTCV. Our proposed EdGE method consistently exhibits excellent performance across various teacher-student combinations. These results suggest that our method holds promise in bridging the knowledge gap with a significant margin compared to other competitive methods. The effectiveness of the individual modules is further confirmed via ablation experiments and visualization results.

关键词： Knowledge distillation Knowledge transfer degradation Heterogeneous networks Edge-sensitive knowledge 3d medical image segmentation

来源：评论

学校读者我要写书评

暂无评论

LW-CTrans: A lightweight hybrid network of CNN and Transformer for 3d medical image segmentation

引用

medical image ANALYSIS 2025年 102卷 103545页

作者： Kuang, Hulin Wang, Yahui Tana, Xianzhen Yang, Jialin Sun, Jiarui Liu, Jin Qiu, Wu Zhang, Jingyang Zhang, Jiulou Yang, Chunfeng Wang, Jianxin Chen, Yang Cent South Univ Sch Comp Sci & Engn Hunan Prov Key Lab Bioinformat Changsha 410000 Peoples R China Southeast Univ Sch Comp Sci & Engn Nanjing 210096 Peoples R China Southeast Univ Key Lab New Generat Artificial Intelligence Techno Minist Educ Nanjing 210096 Peoples R China Huazhong Univ Sci & Technol Sch Life Sci & Technol Wuhan 430000 Peoples R China Nanjing Med Univ Affiliated Hosp 1 Dept Radiol Nanjing 210096 Peoples R China Nanjing Med Univ Sch Med Imaging Lab Artificial Intelligence Med Imaging LAIMI Nanjing 210096 Peoples R China Xinjiang Univ Xinjiang Engn Res Ctr Big Data & Intelligent Softw Sch Software Urumqi 830091 Xinjiang Peoples R China

Recent models based on convolutional neural network (CNN) and Transformer have achieved the promising performance for 3d medical image segmentation. However, these methods cannot segment small targets well even when equipping large parameters. Therefore, We design a novel lightweight hybrid network that combines the strengths of CNN and Transformers (LW-CTrans) and can boost the global and local representation capability at different stages. Specifically, we first design a dynamic stem that can accommodate images of various resolutions. In the first stage of the hybrid encoder, to capture local features with fewer parameters, we propose a multi-path convolution (MPConv) block. In the middle stages of the hybrid encoder, to learn global and local features meantime, we propose a multi-view pooling based Transformer (MVPFormer) which projects the 3d feature map onto three 2d subspaces to deal with small objects, and use the MPConv block for enhancing local representation learning. In the final stage, to mostly capture global features, only the proposed MVPFormer is used. Finally, to reduce the parameters of the decoder, we propose a multi-stage feature fusion module. Extensive experiments on 3 public datasets for three tasks: stroke lesion segmentation, pancreas cancer segmentation and brain tumor segmentation, show that the proposed LW-CTrans achieves dices of 62.35 +/- 19.51%, 64.69 +/- 20.58% and 83.75 +/- 15.77% on the 3 datasets, respectively, outperforming 16 state-of-the-art methods, and the numbers of parameters (2.08M, 2.14M and 2.21M on 3 datasets, respectively) are smaller than the non-lightweight 3d methods and close to the lightweight methods. Besides, LW-CTrans also achieves the best performance for small lesion segmentation.

关键词： Lightweight network Convolutional neural network and transformer 3d medical image segmentation Small lesion segmentation

来源：评论

学校读者我要写书评

暂无评论

HEdN: multi-oriented hierarchical extraction and dual-frequency decoupling network for 3d medical image segmentation

引用

medical & BIOLOGICAL ENGINEERING & COMPUTING 2025年第1期63卷 267-291页

作者： Wang, Yu Huang, Guoheng Lu, Zeng Wang, Ying Chen, Xuhang Yuan, Xiaochen Li, Yan Ni, Liujie Huang, Yingping Hunan Tradit Chinese Med Coll Publ Courses Dept Zhuzhou 412012 Hunan Peoples R China Guangdong Univ Technol Sch Comp Sci Guangzhou 510006 Guangdong Peoples R China Guangzhou Interesting Pill Network Technol Co Ltd Guangzhou 510630 Guangdong Peoples R China Macao Polytech Univ Fac Appl Sci Taipa 999078 Peoples R China Huizhou Univ Sch Comp Sci & Engn Huizhou 516001 Guangdong Peoples R China Shenzhen Polytech Univ Shenzhen 518000 Guangdong Peoples R China Ningxiang Tradit Chinese Med Hosp Dept Cardiol 8 Second Ring South Rd Ningxiang 410699 Hunan Peoples R China Sun Yat sen Univ Collaborat Innovat Ctr Canc MedCanc Ctr Dept Radiat OncolGuangdong Key Lab Nasopharyngeal State Key Lab Oncol South China Guangzhou 510006 Guangdong Peoples R China

3d encoder-decoder segmentation architectures struggled with fine-grained feature decomposition, resulting in unclear feature hierarchies when fused across layers. Furthermore, the blurred nature of contour boundaries in medical imaging limits the focus on high-frequency contour features. To address these challenges, we propose a Multi-oriented Hierarchical Extraction and dual-frequency decoupling Network (HEdN), which consists of three modules: Encoder-decoder Module (E-dM), Multi-oriented Hierarchical Extraction Module (Multi-HEM), and dual-frequency decoupling Module (dual-dM). The E-dM performs the basic encoding and decoding tasks, while Multi-HEM decomposes and fuses spatial and slice-level features in 3d, enriching the feature hierarchy by weighting them through 3d fusion. dual-dM separates high-frequency features from the reconstructed network using self-supervision. Finally, the self-supervised high-frequency features separated by dual-dM are inserted into the process following Multi-HEM, enhancing interactions and complementarities between contour features and hierarchical features, thereby mutually reinforcing both aspects. On the Synapse dataset, HEdN outperforms existing methods, boosting dice Similarity Score (dSC) by 1.38% and decreasing 95% Hausdorff distance (Hd95) by 1.03 mm. Likewise, on the Automatic Cardiac diagnosis Challenge (ACdC) dataset, HEdN achieves 0.5% performance gains across all categories.

关键词： 3d medical image segmentation Encoder-decoder architecture Multi-oriented hierarchical extraction dual-frequency decoupling

来源：评论

学校读者我要写书评

暂无评论

PFormer: An efficient CNN-Transformer hybrid network with content-driven P-attention for 3d medical image segmentation

引用

BIOmedical SIGNAL PROCESSING ANd CONTROL 2025年 101卷

作者： Gao, Yueyang Zhang, Jinhui Wei, Siyi Li, Zheng Beijing Inst Technol Sch Automat Beijing 100081 Peoples R China

medical imaging, particularly medical image segmentation, is pivotal in modern medicine. Transformer architectures have gained significant attention in the field of medical image segmentation. Both pure transformer architectures and hybrid architectures combining transformer with CNN have been extensively proposed, demonstrating impressive performance in medical image segmentation. However, whether embedding the transformer in the higher layers of CNN networks or enhancing attention mechanisms, these approaches have inherent limitations in fully harnessing the potential of the transformer within the constraints of low computational cost. This paper introduces an efficient hybrid network structure that combines CNN and transformer for 3d medical image segmentation. This network utilizes depth-wise separable convolutions (dWconvs) to extract local fine-grained details and introduces a content-driven attention mechanism, Pattention, to replace self-attention within the transformer. P-attention enables global feature modeling in both channel and spatial dimensions. By integrating local information with global dependencies, the network achieves more precise segmentation results with a reduced computational burden. Extensive experiments on Synapse and Automated Cardiac diagnosis Challenge (ACdC) datasets, utilizing various evaluation metrics such as dice similarity coefficients (dSC), 95% percentile Hausdorff distance (Hd95), average symmetric surface distance (ASSd), precision (PRE) and sensitivity (SE), were conducted to assess the performance of PFormer. The experimental results demonstrate that, compared to other state-of-the-art methods, PFormer achieves superior segmentation performance under lower parameters and floating point of operations (FLOPs), providing strong evidence of the effectiveness of our method. Codes and models of PFormer are available at https://***/BitGyy/PFormer.

关键词： 3d medical image segmentation CNN-Transformer hybrid architecture P-attention Various evaluation metrics Superior segmentation performance Low computational cost

来源：评论

学校读者我要写书评

暂无评论

A model use context complementarity feature fusion learning for semi-supervised 3d medical image segmentation

引用

BIOmedical SIGNAL PROCESSING ANd CONTROL 2025年 102卷

作者： Chen, Lei Zhao, Yikai Yang, dongxu Ma, Yunpeng Zhao, Bingjie Hou, Jieru Liu, Wenhao Tianjin Univ Commerce Sch Informat Engn Tianjin 300134 Peoples R China

In 3d medical image segmentation, Semi-Supervised Learning (SSL) methods have shown strong potential with limited labeled data. However, most existing SSL models fail to effectively utilize correlations between data slices, leading to suboptimal segmentation boundaries. This paper presents a novel model for semi-supervised 3d medical image segmentation, namely, context complementary feature fusion learning (CCFFL). Specifically, we designed a novel encoder-decoder network and proposed context complementary and feature fusion learning modules. The context complementary module is embedded in the encoder-decoder structure. Rich spatial features are extracted in the encoding stage to avoid information loss, and high-resolution images are recovered in the decoding stage to improve the segmentation accuracy and the utilization efficiency of contextual information. Moreover, the feature fusion learning module evaluates, fuses, and enhances features extracted from CT and MRI slices and generates medical feature maps with global information. We assessed the CCFFL model in the left atrium and pancreas datasets and validated the feature mining capability of CCFFL in a series of ablation experiments. Experimental results show that our model outperforms most semi-supervised medical image segmentation models, with dice scores of 90.09 % and 77.76 % in the left atrium and pancreas, respectively.

关键词： Semi-supervised learning 3d medical image segmentation data slices Encoder-decoder network Context complementary module Feature fusion learning module

来源：评论

学校读者我要写书评

暂无评论

Self-supervised 3d medical image segmentation by flow-guided mask propagation learning

引用

medical image ANALYSIS 2025年 101卷 103478页

作者： Bitarafan, Adeleh Mozafari, Mohammad Azampour, Mohammad Farid Baghshah, Mahdieh Soleymani Navab, Nassir Farshad, Azade Sharif Univ Technol Tehran Iran Tech Univ Munich Comp Aided Med Procedures Munich Germany Munich Ctr Machine Learning Munich Germany

despite significant progress in 3d medical image segmentation using deep learning, manual annotation remains a labor-intensive bottleneck. Self-supervised mask propagation (SMP) methods have emerged to alleviate this challenge, allowing intra-volume segmentation with just a single slice annotation. However, the previous SMP methods often rely on 2d information and ignore volumetric contexts. While our previous work, called Vol2Flow, attempts to address this concern, it exhibits limitations, including not focusing enough on local (i.e., slice-pair) information, neglecting global information (i.e., volumetric contexts) in the objective function, and error accumulation during slice-to-slice reconstruction. This paper introduces Flow2Mask, a novel SMP method, developed to overcome the limitations of previous SMP approaches, particularly Vol2Flow. during training, Flow2Mask proposes the Local-to-Global (L2G) loss to learn inter-slice flow fields among all consecutive slices within a volume in an unsupervised manner. This dynamic loss is based on curriculum learning to gradually learn information within a volume from local to global contexts. Additionally, the Inter-Slice Smoothness (ISS) loss is introduced as a regularization term to encourage changes between the slices occur consistently and continuously. during inference, Flow2Mask leverages these 3d flow fields for inter-slice mask propagation in a 3d image, spreading annotation from a single annotated slice to the entire volume. Moreover, we propose an automatic strategy to select the most representative slice as initial annotation in the mask propagation process. Experimental evaluations on different abdominal datasets demonstrate that our proposed SMP method outperforms previous approaches and improves the overall mean dSC of Vol2Flow by +2.1%, +8.2%, and +4.0% for the Sliver, CHAOS, and 3d-IRCAd datasets, respectively. Furthermore, Flow2Mask even exhibits substantial improvements in weakly-supervised and self-supe

关键词： 3d medical image segmentation Mask propagation Self-supervised learning Sparse annotation image registration

来源：评论

学校读者我要写书评

暂无评论

Attention correction feature and boundary constraint knowledge distillation for efficient 3d medical image segmentation

引用

EXPERT SYSTEMS WITH APPLICATIONS 2025年 262卷

作者： Yu, Xiangchun Teng, Longxiang Zhang, dingwen Zheng, Jian Chen, Hechang Jiangxi Univ Sci & Technol Sch Informat Engn Jiangxi Prov Key Lab Multidimens Intelligent Perce Ganzhou 341000 Peoples R China Jilin Univ Sch Artificial Intelligence Changchun 130012 Peoples R China

Incorporating auxiliary modules into the teacher to facilitate feature knowledge transfer often incurs additional training costs, significantly impeding the application of feature-based knowledge distillation in 3d medical image segmentation. This paper explores the non-negative matrix factorization (NMF) theory to illustrate that ResNet-like or UNet-like residual connections and attention mechanisms function as corrections to input features. Based on this discovery, we propose Attention Correction Feature distillation (ACFd), which comprises Correction Factor distillation (CFd) and Attention-based Feature distillation (AFd). In CFd, the teacher's residual correction factor guides the student's learning of the attention correction factor, while in AFd, the teacher's residual-corrected features are used to teach the student's attention-corrected features. Furthermore, the introduction of boundary constraints can enhance the student's boundary perception, but Hard Boundary distillation (Hard Bd) is sensitive to noise. Hence, we introduce Soft Boundary distillation (Soft Bd) to extract soft boundary information by estimating the Hausdorff distance (Hd) and integrate both Hard Bd and Soft Bd into Boundary Constraint distillation (BCd). Ultimately, we present the Attention Correction Feature and Boundary Constraint Knowledge distillation (ACF-BCKd). Comprehensive experiments on three benchmark datasets - BTCV, WORd, and BraTS-23 - demonstrate the state-of-the-art performance of ACF-BCKd. Ablation experiments and visualization results confirm the effectiveness of individual modules.

关键词： Feature-based knowledge distillation 3d medical image segmentation Non-negative matrix factorization Correction factor Boundary constraint distillation

来源：评论

学校读者我要写书评

暂无评论

3d medical image segmentation based on semi-supervised learning using deep co-training

引用

APPLIEd SOFT COMPUTING 2024年 159卷

作者： Yang, Jingdong Li, Haoqiu Wang, Han Han, Man Univ Shanghai Sci & Technol Sch Opt Elect & Comp Engn Shanghai 200093 Peoples R China China Acad Chinese Med Sci Guanganmen Hosp Div Rheumatol Beijing 100053 Peoples R China

In recent years, artificial intelligence has been applied to 3d COVId-19 medical image diagnosis, which reduces detection costs and missed diagnosis rates with higher predictive accuracy, and diagnostic efficiency. However, the limited size and low quality of clinical 3d medical image samples have hindered the segmentation performance of 3d models. Therefore, we propose a 3d medical image segmentation model based on semi-supervised learning using co-training. Multi-view and multi-modal images are generated using spatial flipping and windowing techniques to enhance the spatial diversity of 3d image samples. A pseudo label generation module based on confidence-weights is employed to generate reliable pseudo labels for non-annotated data, thereby increasing the sample size and reducing overfitting. The proposed approach utilizes a three-stage training process: firstly, training a single network based on annotated data;secondly, incorporating non-annotated data to train a dualmodal network and generate pseudo labels;finally, jointly training six models in three dimensions using both annotated and pseudo labels generated from multi-view and multi-modal images, aiming to enhance segmentation accuracy and generalization performance. Additionally, a consistency regularization loss is applied to reduce noises and accelerate convergence of the training. Moreover, a heatmap visualization method is employed to focus on the attention of features at each stage of training, providing effective reference for clinical diagnosis. Experiments were conducted on an open dataset of 3d COVId-19 CT samples and a non-annotated dataset from TCIA, including 771 NIFTI-format CT images from 661 COVId-19 patients. The results of 5-fold cross-validation show that the proposed model achieves a segmentation accuracy of dice=73.30 %, ASd=10.633, Sensitivity=63.00 %, and Specificity=99.60 %. Compared to various typical semi-supervised learning 3d segmentation models, it demonstrates better segmentati

关键词： 3d medical image segmentation Semi -supervised learning Co -training Pseudo labels

来源：评论

学校读者我要写书评

暂无评论

dS-Former: A dual-stream encoding-based transformer for 3d medical image segmentation

引用

BIOmedical SIGNAL PROCESSING ANd CONTROL 2024年 89卷

作者： Zhang, Lei Zuo, Yi Jia, Yu Li, dongze Zeng, Rui Li, dong Chen, Junren Wang, Wei Sichuan Univ Comp Sci 24 South Sect 1Yihuan Rd Chengdu 610065 Sichuan Peoples R China Sichuan Univ West China Hosp Int Med Ctr Ward Gen Practice Med CtrGen Practice Ward 37 Guoxue Alley Chengdu 610065 Sichuan Peoples R China Sichuan Univ West China Hosp West China Sch Med Dept Emergency MedDisaster Med Ctr 37 Guoxue Alley Chengdu 610065 Sichuan Peoples R China Sichuan Univ West China Hosp West China Sch Med Dept Cardiol 37 Guoxue Alley Chengdu 610065 Sichuan Peoples R China Qinghai Univ Dept Comp Sci 251 Ningda Rd Xining 810016 Qinghai Peoples R China Chengdu Univ Informat Technol Sch Automat 24 Block 1Xuefu Rd Chengdu 610225 Sichuan Peoples R China

Models that utilize self-attention mechanisms, including but not limited to Vision Transformers (ViTs), have shown promising performance in visual tasks like semantic segmentation. This is attributed to their capacity to capture global features of images, enabling them to learn more comprehensive representations. However, transformer-based models typically demand a considerable amount of training data to achieve satisfactory performance, while being deficient in the ability to efficiently extract local image features. As a result, these models may not be as effective in some computer vision tasks that involve small-scale datasets, like medical image segmentation. To address these issues, this paper proposes a dual-stream encoding-based transformer dubbed as dual-stream Transformer (dS-Former). The dual-stream module in dS-Former can simultaneously acquire local and global features in the image and construct relation between the two kinds of features via self-attention. Compared with the simple splicing or serial connection, the dual-stream module can extract more comprehensive and hierarchical feature information from the fusion interaction of the two features. Our method is evaluated on the UK Biobank (UKBB) cardiac magnetic resonance imaging (CMR) dataset and The Beyond the Cranial Vault (BTCV) abdominal challenge dataset. The experimental results indicate that our dS-Former outperforms other state-of-the-art approaches on both datasets, indicating its potential for medical images semantic segmentation.

关键词： dual-stream module Transformer Convolution 3d medical image segmentation

来源：评论

学校读者我要写书评

暂无评论

TPAFNet: Transformer-driven Pyramid Attention Fusion Network for 3d medical image segmentation

引用

IEEE JOURNAL OF BIOmedical ANd HEALTH INFORMATICS 2024年第11期28卷 6803-6814页

作者： Li, Zheng Zhang, Jinhui Wei, Siyi Gao, Yueyang Cao, Chengwei Wu, Zhiwei Beijing Inst Technol Sch Automat Beijing 100081 Peoples R China

The field of 3d medical image segmentation is witnessing a growing trend in the utilization of combined networks that integrate convolutional neural networks and transformers. Nevertheless, prevailing hybrid networks are confronted with limitations in their straightforward serial or parallel combination methods and lack an effective mechanism to fuse channel and spatial feature attention. To address these limitations, we present a robust multi-scale 3d medical image segmentation network, the Transformer-driven Pyramid Attention Fusion Network, which is denoted as TPAFNet, leveraging a hybrid structure of CNN and transformer. Within this framework, we exploit the characteristics of atrous convolution to extract multi-scale information effectively, thereby enhancing the encoding results of the transformer. Furthermore, we introduce the TPAF block in the encoder to seamlessly fuse channel and spatial feature attention from multi-scale feature inputs. In contrast to conventional skip connections that simply concatenate or add features, our decoder is enriched with a TPAF connection, elevating the integration of feature attention between low-level and high-level features. Additionally, we propose a low-level encoding shortcut from the original input to the decoder output, preserving more original image features and contributing to enhanced results. Finally, the deep supervision is implemented using a novel CNN-based voxel-wise classifier to facilitate better network convergence. Experimental results demonstrate that TPAFNet significantly outperforms other state-of-the-art networks on two public datasets, indicating that our research can effectively improve the accuracy of medical image segmentation, thereby assisting doctors in making more precise diagnoses.

关键词： Transformers image segmentation Feature extraction Three-dimensional displays decoding Convolution medical diagnostic imaging 3d medical image segmentation transformer attention fusion mechanism

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：