| 范晓冬,张楚浩,周婧,侯利民,韩莹.卷积和自注意力机制双融合的医学图像分割网络[J].电子测量与仪器学报,2026,40(5):156-165 |
| 卷积和自注意力机制双融合的医学图像分割网络 |
| Medical image segmentation network based on dual fusion ofconvolutions and self-attention mechanisms |
| |
| DOI： |
| 中文关键词: 医学图像分割 卷积神经网络 Transformer 可变形卷积 深度可分离卷积 |
| 英文关键词:medical image segmentation convolutional neural network Transformer deformable convolution depthwise separable convolution |
| 基金项目:国家自然科学基金（52177047，62203197）项目资助 |
|
|
| Author | Institution |
| Fan Xiaodong | College of Electrical and Control Engineering, Liaoning Technical University, Huludao 125105，China |
| Zhang Chuhao | College of Electrical and Control Engineering, Liaoning Technical University, Huludao 125105，China |
| Zhou Jing | College of Mathematical Sciences, Bohai University, Jinzhou 121000,China |
| Hou Limin | College of Electrical and Control Engineering, Liaoning Technical University, Huludao 125105，China |
| Han Ying | College of Electrical and Control Engineering, Liaoning Technical University, Huludao 125105，China |
|
| 摘要点击次数: 243 |
| 全文下载次数: 72 |
| 中文摘要: |
| 针对医学图像分割中局部纹理细节与全局语义信息难以充分融合、易造成边界模糊与结构不完整的问题，设计了一种卷积与自注意力并行的双融合分支医学图像分割U形网络 (dual fusion branch UNet，DFB-UNet)。编码器采用卷积神经网络 (convolutional neural network，CNN)分支与Transformer分支并行提取局部与全局特征；在跨分支融合前端引入可变形卷积网络(deformable convolution network，DCN)以增强对不规则目标边界的建模能力；并构建由注意力融合模块(attention fusion module，AFM)与卷积融合模块 (convolution fusion module，CFM)组成的双融合模块，其中CFM引入深度可分离卷积以在较低计算开销下实现有效特征交互与互补聚合。实验结果显示，在Kvasir-SEG数据集上mDice/mIoU达到0.89/0.83；在CVC-ClinicDB数据集上mDice/mIoU为0.810/0.723；跨数据集测试中，使用Kvasir-SEG训练并在CVC-ClinicDB测试时Dice/IoU为0.751/0.668；在ISIC2017数据集上IoU/Dice/Acc为0.816/0.889/0.965，在LUNA16数据集上Dice/IoU为0.966 0/0.934 2。结果表明，所提出的双融合并行编码结构能够在保证计算效率的同时提升分割精度与泛化性能，消融实验进一步验证了可变形卷积与双融合模块对网络性能提升的贡献。 |
| 英文摘要: |
| To address the insufficient fusion of local texture details and global semantic information in medical image segmentation, which often leads to blurred boundaries and incomplete structures, a named dual fusion branch UNet（DFB-UNet）is proposed with parallel convolution and self-attention. The encoder employs a convolutional neural network（CNN）branch and a Transformer branch in parallel to learn local and global features. Before cross-branch fusion, a deformable convolution network （DCN）is introduced to enhance the modeling of irregular object boundaries. In addition, a dual-fusion module composed of an attention fusion module（AFM）and a convolution fusion module（CFM）is constructed; CFM incorporates depthwise separable convolutions to achieve effective feature interaction and complementary aggregation with lower computational cost. The experiment results are as follows: mDice/mIoU reach 0.89/0.83 on the dataset Kvasir-SEG; 0.810/0.723 on CVC-ClinicDB. In cross-dataset testing, training on Kvasir-SEG and testing on CVC-ClinicDB yield Dice/IoU of 0.751/0.668. On ISIC2017, IoU/Dice/Acc are 0.816/0.889/0.965, and on LUNA16, Dice/IoU are 0.966 0/0.934 2. The results indicate that the proposed dual-fusion parallel encoder improves segmentation accuracy and generalization while maintaining computational efficiency. The ablation studies further verify the contribution of DCN and the dual-fusion module. |
| 查看全文 查看/发表评论 下载PDF阅读器 |
|
|
|
