基于多尺度特征融合和注意力机制的医学图像分割网络

Medical image segmentation network based on multi-scale feature fusion and attention mechanism

王龙业 ¹张凯信 ¹曾晓莉 ²方东 ¹李沁 ¹马傲¹

扫码查看

作者信息

1. 西南石油大学电气信息学院,四川成都 610500
2. 西藏大学信息科学技术学院,西藏拉萨 850000
折叠

摘要

针对传统编解码结构的医学图像分割网络存在特征信息利用率低、泛化能力不足等问题,该文提出了一种结合编解码模式的多尺度语义感知注意力网络(multi-scale semantic perceptual at-tention network,MSPA-Net).首先,该网络在解码路径加入双路径多信息域注意力模块(dual-channel multi-information domain attention module,DMDA),提高特征信息的提取能力;其次,网络在级联处加入空洞卷积模块(dense atrous convolution module,DAC),扩大卷积感受野;最后,借鉴特征融合思想,设计了可调节多尺度特征融合模块(adjustable multi-scale feature fusion,AMFF)和双路自学习循环连接模块(dual self-learning recycle connection module,DCM),提升网络的泛化性和鲁棒性.为验证网络的有效性,在 CVC-ClinicDB、ETIS-LaribPolypDB、COVID-19 CHEST X-RAY、Kaggle_3m、ISIC2017和Fluorescent Neuronal Cells等数据集上进行验证,实验结果表明,相似系数分别达到了94.96％、92.40％、99.02％、90.55％、92.32％和75.32％.因此,新的分割网络展现了良好的泛化能力,总体性能优于现有网络,能够较好实现通用医学图像的有效分割.

Abstract

Aiming at the problems of low utilization of feature information and insufficient generalization ability in the traditional medical image segmentation network with encoding and decoding structure,this paper proposes a multi-scale semantic perceptual attention network(MSPA-Net)combined with enco-ding and decoding mode.Firstly,the network adds a dual-channel multi-information domain attention module(DMDA)to the decoding path to improve the ability of feature information extraction.Secondly,the network adds a dense atrous convolution module(DAC)at the cascade to expand the convolution re-ceptive field.Finally,based on the idea of feature fusion,an adjustable multi-scale features fusion module(AMFF)and a dual self-learning recycle connection module(DCM)are designed to improve the gener-alization and robustness of the network.To verify the effectiveness of the network,the experimental ver-ification is carried out on CVC-ClinicDB,ETIS-LaribPolypDB,COVID-19 CHEST X-RAY,Kaggle_3m,ISIC 2017,and Fluorescent Neuronal Cells datasets,and the similarity coefficients reach 94.96％,92.40％,99.02％,90.55％,92.32％and 75.32％respectively.Therefore,the new segmentation net-work shows better generalization ability,the overall performance is better than the existing network,and can better achieve the effective segmentation of general medical images.

关键词

医学图像分割/注意力机制/特征融合/空洞卷积

Key words

medical image segmentation/attention mechanism/feature fusion/atrous convolution

引用本文复制引用

基金项目

国家自然科学基金(61261021)

国家自然科学基金(61561045)

四川省科技计划项目(2019JDRC0012)

出版年

2024

光电子·激光

天津理工大学中国光学学会

光电子·激光

CSCD北大核心

影响因子：1.437

ISSN：1005-0086

参考文献量32

段落导航