重庆理工大学学报(自然科学) ›› 2023, Vol. 37 ›› Issue (5): 185-193.

• 信息·计算机 • 上一篇    下一篇

基于卷积融合和残差注意力的脑卒中病灶分割

张 岩,李凤莲,张雪英,王夙?,章洪涛   

  1. (太原理工大学 信息与计算机学院,太原 030024)
  • 出版日期:2023-06-21 发布日期:2023-06-21
  • 作者简介:张岩,男,硕士研究生,主要从事医学图像分割、深度学习研究,Email:zhangyan0414@link.tyut.edu.cn;通信作 者 李凤莲,女,博士,教授,主要从事医疗脑卒中信号处理、非平衡数据集分析及应用研究,Email:ghllfl@ 163.com。

Stroke lesion segmentation based on convolution fusion and residual-attention mechanism

  • Online:2023-06-21 Published:2023-06-21

摘要: 脑卒中 MRI影像由于病灶区域小和正常组织边界模糊的特点导致分割难度大。 为此提出一种优化的编解码结构网络。为使网络提取更加丰富的上下文信息,提出了双注意力 卷积融合编码模块,在编码端收缩路径实现二维卷积和三维卷积的融合,并且从空间和通道 2 个维度建立特征的全局相关性。此外,提出残差注意力门混合解码模块,更好地融合低层次和 高层次特征,关注目标区域,从而提高小病灶边缘的分割细腻度。通过在开源数据集 ATLAS的 实验结果表明,该算法 DSC指标达到了0.62,与 UNet,DUNet,3DUNet以及 attentionUNet等模 型相比,有效提高了分割性能。

关键词: 脑卒中, UNet, 卷积融合, 注意力

Abstract: Stroke MRI images are difficult to be segmented due to small lesion areas and blurred boundaries between lesions and healthy tissues. Therefore, this paper proposes an optimized encoder-decoder structure network. Firstly, in order to extract richer contextual information for the network, a dual-attention convolution fusion coding module is established, which combines 2D convolution with 3D convolution on the contraction path of the encoding end, and builds global correlation of the features from both space and channel dimensions. In addition, a residual-attention gate hybrid decoding module is proposed to better fuse low-level and high-level features and focus on the target area, thereby improving the segmentation fineness of the edges of the small lesions. The experimental results in the open source dataset anatomical tracings of lesions after stroke (ATLAS) show that the DSC index of the algorithm reaches 0.62. Compared with models such as UNet, D-UNet, 3D-UNet and attention-UNet, it demonstrates a much improved segmentation performance.

中图分类号: 

  • TP391