FSOU-Net: Feature supplement and optimization U-Net for 2D medical image segmentation

被引:2
|
作者
Wang, Yongtao [1 ,2 ]
Tian, Shengwei [1 ,2 ]
Yu, Long [3 ]
Wu, Weidong [4 ,5 ]
Zhang, Dezhi [4 ]
Wang, Junwen [1 ,2 ]
Cheng, Junlong [6 ]
机构
[1] Xinjiang Univ, Coll Software Engn, Urumqi, Xinjiang, Peoples R China
[2] Xinjiang Univ, Key Lab Software Engn Technol, Urumqi, Xinjiang, Peoples R China
[3] Xinjiang Univ, Coll Informat Sci & Engn, Urumqi, Xinjiang, Peoples R China
[4] Peoples Hosp Xinjiang Uygur Autonomous Reg, Urumqi, Xinjiang, Peoples R China
[5] China Xinjiang Key Lab Dermatol Res XJYS1707, Urumqi, Xinjiang, Peoples R China
[6] Sichuan Univ, Coll Comp Sci, Chengdu, Sichuan, Peoples R China
关键词
Deep learning algorithm; medical image analysis; semantic segmentation; multi-scale; convolutional neural networks; OPTIC DISC;
D O I
10.3233/THC-220174
中图分类号
R19 [保健组织与事业(卫生事业管理)];
学科分类号
摘要
BACKGROUND: The results of medical image segmentation can provide reliable evidence for clinical diagnosis and treatment. The U-Net proposed previously has been widely used in the field of medical image segmentation. Its encoder extracts semantic features of different scales at different stages, but does not carry out special processing for semantic features of each scale. OBJECTIVE: To improve the feature expression ability and segmentation performance of U-Net, we proposed a feature supplement and optimization U-Net (FSOU-Net). METHODS: First, we put forward the view that semantic features of different scales should be treated differently. Based on this view, we classify the semantic features automatically extracted by encoders into two categories: shallow semantic features and deep semantic features. Then, we propose the shallow feature supplement module (SFSM), which obtains fine-grained semantic features through up-sampling to supplement the shallow semantic information. Finally, we propose the deep feature optimization module (DFOM), which uses the expansive convolution of different receptive fields to obtain multi-scale features and then performs multi-scale feature fusion to optimize the deep semantic information. RESULTS: The proposed model is experimented on three medical image segmentation public datasets, and the experimental results prove the correctness of the proposed idea. The segmentation performance of the model is higher than the advanced models for medical image segmentation. Compared with baseline network U-NET, the main index of Dice index is 0.75% higher on the RITE dataset, 2.3% higher on the Kvasir-SEG dataset, and 0.24% higher on the GlaS dataset. CONCLUSIONS: The proposed method can greatly improve the feature representation ability and segmentation performance of the model.
引用
收藏
页码:181 / 195
页数:15
相关论文
共 50 条
  • [31] A Bypass-Based U-Net for Medical Image Segmentation
    Chen, Kaixuan
    Xu, Gengxin
    Qian, Jiaying
    Ren, Chuan-Xian
    INTELLIGENCE SCIENCE AND BIG DATA ENGINEERING: VISUAL DATA ENGINEERING, PT I, 2019, 11935 : 155 - 164
  • [32] RotU-Net: An Innovative U-Net With Local Rotation for Medical Image Segmentation
    Zhang, Fuxiang
    Wang, Fengchao
    Zhang, Wenfeng
    Wang, Quanzhen
    Liu, Yajun
    Jiang, Zhiming
    IEEE ACCESS, 2024, 12 : 21114 - 21128
  • [33] Comparison of tissue segmentation performance between 2D U-Net and 3D U-Net on brain MR Images
    Woo, Boyeong
    Lee, Myungeun
    2021 INTERNATIONAL CONFERENCE ON ELECTRONICS, INFORMATION, AND COMMUNICATION (ICEIC), 2021,
  • [34] U-Net vs Transformer: Is U-Net Outdated in Medical Image Registration?
    Jia, Xi
    Bartlett, Joseph
    Zhang, Tianyang
    Lu, Wenqi
    Qiu, Zhaowen
    Duan, Jinming
    MACHINE LEARNING IN MEDICAL IMAGING, MLMI 2022, 2022, 13583 : 151 - 160
  • [35] IBA-U-Net: Attentive BConvLSTM U-Net with Redesigned Inception for medical image segmentation
    Chen, Siyuan
    Zou, Yanni
    Liu, Peter X.
    COMPUTERS IN BIOLOGY AND MEDICINE, 2021, 135
  • [36] TBUnet: A Pure Convolutional U-Net Capable of Multifaceted Feature Extraction for Medical Image Segmentation
    LiFang Chen
    Jiawei Li
    Hongze Ge
    Journal of Medical Systems, 47
  • [37] 3D U2-Net: A 3D Universal U-Net for Multi-domain Medical Image Segmentation
    Huang, Chao
    Han, Hu
    Yao, Qingsong
    Zhu, Shankuan
    Zhou, S. Kevin
    MEDICAL IMAGE COMPUTING AND COMPUTER ASSISTED INTERVENTION - MICCAI 2019, PT II, 2019, 11765 : 291 - 299
  • [38] TBUnet: A Pure Convolutional U-Net Capable of Multifaceted Feature Extraction for Medical Image Segmentation
    Chen, LiFang
    Li, Jiawei
    Ge, Hongze
    JOURNAL OF MEDICAL SYSTEMS, 2023, 47 (01)
  • [39] Medical image segmentation method based on multi-scale feature and U-net network
    Wang, Jingquan
    INTERNET TECHNOLOGY LETTERS, 2023, 7 (05)
  • [40] Improved Medical Image Segmentation Model Based on 3D U-Net
    林威
    范红
    胡晨熙
    杨宜
    禹素萍
    倪林
    Journal of Donghua University(English Edition), 2022, (04) : 311 - 316