| 注册
首页|期刊导航|光学精密工程|双路径编码和跨级解耦的视网膜血管分割

双路径编码和跨级解耦的视网膜血管分割

张蓉蓉 吕晓琪 李菁 谷宇 刘传强

光学精密工程2025,Vol.33Issue(14):2242-2261,20.
光学精密工程2025,Vol.33Issue(14):2242-2261,20.DOI:10.37188/OPE.20253314.2242

双路径编码和跨级解耦的视网膜血管分割

Dual-path encoding and cross-level decoupling for retinal vessel segmentation

张蓉蓉 1吕晓琪 2李菁 1谷宇 1刘传强1

作者信息

  • 1. 内蒙古科技大学数智产业学院,内蒙古包头 014010
  • 2. 内蒙古科技大学数智产业学院,内蒙古包头 014010||内蒙古工业大学信息工程学院,内蒙古呼和浩特 010051
  • 折叠

摘要

Abstract

Retinal vessel segmentation is a critical foundation for ophthalmic disease diagnosis.However,existing methods have issues such as missed detection of thin vessels,interference from pathological re-gions,and feature entanglement.Therefore,this paper proposed a Dual-path Feature Extraction and Cross-level Feature Decoupling Network.First,in the encoder,a Multi-directional Differential Residual Block extracted vessel-edge features across multiple directions to capture subtle vessel structures.Next,a Collaborative Attention Fusion Block dynamically integrated the complementary convolutional features of the double-branch path,enhancing the aggregation of encoded information.Finally,for the skip connec-tions between the two U-Net branches,a Channel Interaction Disentanglement Block was introduced to decouple cross-level features,strengthen feature representations,and resolve the feature-confusion prob-lem inherent in traditional U-Net architectures.The proposed method was extensively validated on four public datasets—DRIVE,CHASEDB1,STARE,and IOSTAR—achieving F1-scores of 82.47%,80.71%,81.44%,and 82.01%,and sensitivities of 80.96%,80.23%,74.69%,and 76.92%,respectively.The F1 scores of the LadderNet algorithm are 81.66%,80.16%,80.92%and 79.69%,and the sensitivity is 77.06%,78.88%,73.64%and 71.24%,respectively.Compared to state-of-the-art methods,our approach shows good robustness and superior segmentation performance in the task of reti-nal vascular segmentation.

关键词

视网膜血管/图像分割/多向差分卷积/动态特征融合/注意力机制

Key words

retinal vessel/image segmentation/multi-directional differential convolution/dynamic fea-ture fusion/attention mechanism

分类

信息技术与安全科学

引用本文复制引用

张蓉蓉,吕晓琪,李菁,谷宇,刘传强..双路径编码和跨级解耦的视网膜血管分割[J].光学精密工程,2025,33(14):2242-2261,20.

基金项目

国家自然科学基金(No.62001255,No.61771266) (No.62001255,No.61771266)

中央引导地方科技发展资金项目(No.2021ZY0004) (No.2021ZY0004)

内蒙古自治区自然科学基金(No.2024MS06008) (No.2024MS06008)

内蒙古自治区高等学校青年科技英才计划项目(No.NJYT23057) (No.NJYT23057)

内蒙古自治区直属高校基本科研业务费项目优秀青年基金(No.042) (No.042)

光学精密工程

OA北大核心

1004-924X

访问量0
|
下载量0
段落导航相关论文