JOINT ATTENTION FOR MEDICAL IMAGE SEGMENTATION

被引:0
作者
Zhang, Mo [1 ,2 ,3 ]
Dong, Bin [1 ,4 ]
Li, Quanzheng [5 ]
机构
[1] Peking Univ, Ctr Data Sci, Beijing, Peoples R China
[2] Peking Univ, Ctr Data Sci Hlth & Med, Beijing, Peoples R China
[3] Beijing Inst Big Data Res, Lab Biomed Image Anal, Beijing, Peoples R China
[4] Peking Univ, Beijing Int Ctr Math Res BICMR, Beijing, Peoples R China
[5] Massachusetts Gen Hosp, Harvard Med Sch, MGH BWH Ctr Clin Data Sci, Ctr Adv Med Comp & Anal,Dept Radiol, Boston, MA 02114 USA
来源
2022 IEEE INTERNATIONAL SYMPOSIUM ON BIOMEDICAL IMAGING (IEEE ISBI 2022) | 2022年
关键词
Spatial attention; medical image segmentation; self-attention;
D O I
10.1109/ISBI52829.2022.9761624
中图分类号
R318 [生物医学工程];
学科分类号
0831 ;
摘要
Medical image segmentation is crucial for computer aided diagnosis. In recent years, spatial attention mechanisms have leaded to breakthroughs in the task of image segmentation. In this work, we firstly present a unified formula for spatial attention mechanisms. Within this framework, we find that point-wise attention has better localization while self-attention can learn more global features. Motivated by this observation, we then propose a new joint attention module, which jointly leverages the advantages of point-wise attention and self-attention. Moreover, by integrating joint attention with DenseUNet, we conduct image segmentation experiments on two public datasets. The proposed method outperforms recent state-of-the-art models, verifying the superiority of joint attention. Additionally, ablation studies demonstrate that our joint attention obtains more balanced results compared to the previous point-wise attention and self-attention. The design of joint attention provides a novel insight into understanding spatial attention mechanisms.
引用
收藏
页数:5
相关论文
共 17 条
[1]  
Abhishek Kumar, 2020, ARXIV, DOI DOI 10.48550/ARXIV.2010.13454
[2]   Dual Attention Network for Scene Segmentation [J].
Fu, Jun ;
Liu, Jing ;
Tian, Haijie ;
Li, Yong ;
Bao, Yongjun ;
Fang, Zhiwei ;
Lu, Hanqing .
2019 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2019), 2019, :3141-3149
[3]  
Hu J, 2018, PROC CVPR IEEE, P7132, DOI [10.1109/CVPR.2018.00745, 10.1109/TPAMI.2019.2913372]
[4]   Densely Connected Convolutional Networks [J].
Huang, Gao ;
Liu, Zhuang ;
van der Maaten, Laurens ;
Weinberger, Kilian Q. .
30TH IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2017), 2017, :2261-2269
[5]  
Kaul C, 2021, Arxiv, DOI arXiv:1912.02079
[6]   Illumination-based Transformations Improve Skin Lesion Segmentation in Dermoscopic Images [J].
Kumar, Abhishek ;
Hamarneh, Ghassan ;
Drew, Mark S. .
2020 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION WORKSHOPS (CVPRW 2020), 2020, :3132-3141
[7]   H-DenseUNet: Hybrid Densely Connected UNet for Liver and Tumor Segmentation From CT Volumes [J].
Li, Xiaomeng ;
Chen, Hao ;
Qi, Xiaojuan ;
Dou, Qi ;
Fu, Chi-Wing ;
Heng, Pheng-Ann .
IEEE TRANSACTIONS ON MEDICAL IMAGING, 2018, 37 (12) :2663-2674
[8]  
Mo Zhang, 2020, Medical Image Computing and Computer Assisted Intervention - MICCAI 2020. 23rd International Conference. Proceedings. Lecture Notes in Computer Science (LNCS 12264), P321, DOI 10.1007/978-3-030-59719-1_32
[9]  
Oktay O, 2018, Arxiv, DOI [arXiv:1804.03999, 10.48550/arXiv.1804.03999]
[10]  
Park J, 2018, Arxiv, DOI arXiv:1807.06514