SwinT-Unet: Ultrasound Image Segmentation Based on Two-Channel Self-Attention Mechanism

SONG Yan-tao; LU Yun-li

doi:10.12263/DZXB.20230904

您当前的位置：

首页 >

文章列表页 >

SwinT-Unet: Ultrasound Image Segmentation Based on Two-Channel Self-Attention Mechanism

PAPERS | 更新时间：2025-12-11

- SwinT-Unet: Ultrasound Image Segmentation Based on Two-Channel Self-Attention Mechanism
- ACTA ELECTRONICA SINICA Vol. 52, Issue 11, Pages: 3835-3846(2024)
- 作者机构：
  
  1.山西大学大数据科学与产业研究院，山西太原 030006
  2.山西大学计算机与信息技术学院，山西太原 030006
- 作者简介：
- 基金信息：
  
  Shanxi Scholarship Council of China(2023-015)
- DOI：10.12263/DZXB.20230904
  CLC： TP391
- Received：26 September 2023，
  
  Revised：2024-07-24，
  
  Published：25 November 2024
- 稿件说明：
移动端阅览
宋艳涛, 路云里. SwinT-Unet：基于双通道自注意力机制的超声图像分割方法[J]. 电子学报, 2024, 52(11): 3835-3846.

SONG Yan-tao, LU Yun-li. SwinT-Unet: Ultrasound Image Segmentation Based on Two-Channel Self-Attention Mechanism[J]. Acta Electronica Sinica, 2024, 52(11): 3835-3846.
宋艳涛, 路云里. SwinT-Unet：基于双通道自注意力机制的超声图像分割方法[J]. 电子学报, 2024, 52(11): 3835-3846. DOI：10.12263/DZXB.20230904

SONG Yan-tao, LU Yun-li. SwinT-Unet: Ultrasound Image Segmentation Based on Two-Channel Self-Attention Mechanism[J]. Acta Electronica Sinica, 2024, 52(11): 3835-3846. DOI：10.12263/DZXB.20230904

摘要

超声图像分割在疾病诊断和治疗中扮演着关键的角色，但由于超声图像的低对比度、噪声干扰以及病灶在形状、大小和位置上的差异等特点，导致准确地分割出感兴趣的区域仍然是一个具有挑战性的任务.为了解决这一问题，本文提出了一种双通道自注意力机制U型网络（SwinT-Unet），该网络利用Swin-Transformer与Unet编码器同时进行特征提取.为了有效融合Swin-Transformer和Unet编码器提取到的不同层级的特征，本文还提出了一个门控双层特征融合模块（Gated Dual-layer Feature Fusion，GDFF），通过门控机制实现了整体特征与局部特征的有效融合，从而提高分割结果的精确度和鲁棒性.本文在2个不同的超声图像分割数据集上进行了实验，结果表明，本研究所提出的模型在分割准确性和鲁棒性方面均优于现有的卷积神经网络和基于Transformer的网络模型.本文为超声图像分割领域提供了一种新的方法，并为临床医学诊断和治疗提供了更准确、可靠的支持.

Abstract

Ultrasound image segmentation plays a key role in disease diagnosis and treatment

but accurately segmenting the regions of interest is still a challenging task due to the low contrast

noise interference

and variability in shape

size

and location of the lesions in ultrasound images. To address this problem

we propose a dual-channel self-attention mechanism U-shaped network (SwinT-Unet)

which utilizes Swin-Transformer and Unet encoder to simultaneously extract features. To effectively fuse the different-level features extracted by Swin-Transformer and Unet encoder

we also propose a gated dual-layer feature fusion module (GDFF)

which achieves the effective fusion of global and local features through the gating mechanism

thereby improving the accuracy and robustness of the segmentation results. We conduct experiments on two different ultrasound image datasets

and the results show that our proposed model outperforms the existing convolutional neural network and Transformer-based models in terms of segmentation accuracy and robustness. Our paper provides a new method for ultrasound image segmentation

and offers more accurate and reliable support for clinical medical diagnosis and treatment.

关键词

Keywords

references

黄鑫 , 胡艳波 . 国产医疗设备临床应用现状与对策探讨 [J ] . 医疗卫生装备 , 2018 , 39 ( 9 ): 75 - 78 .

HUANG X , HU Y B . Problems and countermeasure of clinical application of domestic medical equipment in China [J ] . Chinese Medical Equipment Journal , 2018 , 39 ( 9 ): 75 - 78 . (in Chinese)

张淑军 , 彭中 , 李辉 . SAU-Net:基于U-Net和自注意力机制的医学图像分割方法 [J ] . 电子学报 , 2022 , 50 ( 10 ): 2433 - 2442 .

ZHANG S J , PENG Z , LI H . SAU-Net: Medical image segmentation method based on U-Net and self-attention [J ] . Acta Electronica Sinica , 2022 , 50 ( 10 ): 2433 - 2442 . (in Chinese)

RONNEBERGER O , FISCHER P , BROX T . U-Net: Convolutional networks for biomedical image segmentation [M ] // Lecture Notes in Computer Science . Cham : Springer International Publishing , 2015 : 234 - 241 .

ALOM M Z , HASAN M , YAKOPCIC C , et al . Recurrent residual convolutional neural network based on U-Net (R2U-Net) for medical image segmentation [EB/OL ] . ( 2018-05-29 )[ 2023-05-24 ] . http://arxiv.org/abs/1802.06955v5 http://arxiv.org/abs/1802.06955v5 .

MILLETARI F , NAVAB N , AHMADI S A . V-net: Fully convolutional neural networks for volumetric medical image segmentation [C ] // 2016 Fourth International Conference on 3D Vision (3DV) . Piscataway : IEEE , 2016 : 565 - 571 .

VASWANI A , SHAZEER N , PARMAR N , et al . Attention is all you need [C ] // Proceedings of the 31st International Conference on Neural Information Processing Systems . Long Beach : Curran Associates Inc , 2017 : 6000 - 6010 .

RAJAMANI K T , RANI P , SIEBERT H , et al . Attention-augmented U-Net (AA-U-Net) for semantic segmentation [J ] . Signal, Image and Video Processing , 2023 , 17 ( 4 ): 981 - 989 .

DOSOVITSKIY A , BEYER L , KOLESNIKOV A , et al . An image is worth 16 x 16 words: Transformers for image recognition at scale[EB/OL ] . ( 2021-06-03 )[ 2023-05-24 ] . http://arxiv.org/abs/2010.11929v2 http://arxiv.org/abs/2010.11929v2 .

CHEN J N , LU Y Y , YU Q H , et al . TransUNet: Transformers make strong encoders for medical image segmentation [EB/OL ] . ( 2021-02-08 )[ 2023-05-24 ] . http://arxiv.org/abs/2102.04306 http://arxiv.org/abs/2102.04306 .

VALANARASU J M J , OZA P , HACIHALILOGLU I , et al . Medical transformer: Gated axial-attention for medical image segmentation [M ] // Lecture Notes in Computer Science . Cham : Springer International Publishing , 2021 : 36 - 46 .

LIU Z , LIN Y T , CAO Y , et al . Swin transformer: Hierarchical vision transformer using shifted windows [C ] // 2021 IEEE/CVF International Conference on Computer Visi-on (ICCV) . Piscataway : IEEE , 2021 : 9992 - 10002 .

CAO H , WANG Y Y , CHEN J , et al . Swin-Unet: Unetlike pure transformer for medical image segmentati-on [M ] // Lecture Notes in Computer Science . Cham : Springer Nature Switzerland , 2023 : 205 - 218 .

TULI S , DASGUPTA I , GRANT E , et al . Are convolutional neural networks or transformers more like human vision? [EB/OL ] . ( 2021-07-01 )[ 2023-05-24 ] . http://arxiv.org/abs/2105.07197v2 http://arxiv.org/abs/2105.07197v2 .

ZHANG Z X , LIU Q J , WANG Y H . Road extraction by deep residual U-net [J ] . IEEE Geoscience and Remote Sensing Letters , 2018 , 15 ( 5 ): 749 - 753 .

ZHOU Z W , SIDDIQUEE M M R , TAJBAKHSH N , et al . UNet++: A nested u-net architecture for medical image segmentation [M ] // Lecture Notes in Computer Science . Cham : Springer International Publishing , 2018 , 11045 : 3 - 11 .

CAI S J , TIAN Y X , LUI H , et al . Dense-UNet: A novel multiphoton in vivo cellular image segmentation model based on a convolutional neural network [J ] . Quantitative Imaging in Medicine and Surgery , 2020 , 10 ( 6 ): 1275 - 1285 .

BACCOUCHE A , GARCIA-ZAPIRAIN B , CASTILLO OLEA C , et al . Connected-UNets: A deep learning architecture for breast mass segmentation [J ] . NPJ Breast Cancer , 2021 , 7 ( 1 ): 151 .

REHMAN M U , CHO S , KIM J H , et al . BU-net: Brain tumor segmentation using modified U-net architecture [J ] . Electronics , 2020 , 9 ( 12 ): 2203 .

WANG X L , GIRSHICK R , GUPTA A , et al . Non-local neural networks [C ] // 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition . Piscataway : IEEE , 2018 : 7794 - 7803 .

OKTAY O , SCHLEMPER J , LE FOLGOC L , et al . Attention U-net: Learning where to look for the pancreas [EB/OL ] . ( 2018-05-20 )[ 2023-05-24 ] . http://arxiv.org/abs/1804.03999v3 http://arxiv.org/abs/1804.03999v3 .

SCHLEMPER J , OKTAY O , CHEN L , et al . Attention-gated networks for improving ultrasound scan plane detection [EB/OL ] . ( 2018-04-15 )[ 2023-05-24 ] . http://arxiv.org/abs/1804.05338 http://arxiv.org/abs/1804.05338 .

PETIT O , THOME N , RAMBOUR C , et al . U-Net transformer: Self and cross attention for medical image segmentation [M ] // Lecture Notes in Computer Science . Cham : Springer International Publishing , 2021 : 267 - 276 .

XIE E Z , WANG W H , YU Z D , et al . SegFormer: Simple and efficient design for semantic segmentation with transformers [EB/OL ] . ( 2021-10-28 )[ 2023-05-24 ] . http://arxiv.org/abs/2105.15203v3 http://arxiv.org/abs/2105.15203v3 .

GONG H F , CHEN G Q , WANG R R , et al . Multi-task learning for thyroid nodule segmentation with thyroid region prior [C ] // 2021 IEEE 18th International Symposium on Biomedical Imaging (ISBI) . Piscataway : IEEE , 2021 : 257 - 261 .

AL-DHABYANI W , GOMAA M , KHALED H , et al . Dataset of breast ultrasound images [J ] . Data in Brief , 2020 , 28 : 104863 .

DU J , GUAN K , LIU P , et al . Boundary-sensitive loss function with location constraint for hard region segmentation [J ] . IEEE Journal of Biomedical and Health Informatics , 2023 , 27 ( 2 ): 992 - 1003 .

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

A Brain Tumor Segmentation Method Based on Softmax Regression and Graph Cut

Enhanced PFCM Algorithm for Noisy Image Segmentation Combining Regional and Neighborhood-Level Information

A Variational Level Set Model Based on Beer-Lambert Law

Medical Image Synthesis Using Robust Conditional GAN

Related Author

GE Ting

MU Ning

LI Li

WANG Xiao-peng

WANG Hai-zhou

CHEN Hao-ran

TANG Li-ming

XIONG Dian-hua

Related Institution

南京理工大学电子工程与光电技术学院

金陵科技学院公共基础课部

School of Electronic and Information Engineering, Lanzhou Jiaotong University

College of Mathematics and Statistics, Hubei Minzu University

⁰