Fundus image segmentation via hierarchical feature learning

被引:21
作者
Guo, Song [1 ]
机构
[1] Xian Univ Architecture & Technol, Sch Informat & Control Engn, Xian 710055, Peoples R China
关键词
High-resolution feature; Hierarchical network; Vessel segmentation; Lesion segmentation; VESSEL SEGMENTATION; RETINAL IMAGES; DIABETIC-RETINOPATHY; NETWORK;
D O I
10.1016/j.compbiomed.2021.104928
中图分类号
Q [生物科学];
学科分类号
07 ; 0710 ; 09 ;
摘要
Fundus Image Segmentation (FIS) is an essential procedure for the automated diagnosis of ophthalmic diseases. Recently, deep fully convolutional networks have been widely used for FIS with state-of-the-art performance. The representative deep model is the U-Net, which follows an encoder-decoder architecture. I believe it is suboptimal for FIS because consecutive pooling operations in the encoder lead to low-resolution representation and loss of detailed spatial information, which is particularly important for the segmentation of tiny vessels and lesions. Motivated by this, a high-resolution hierarchical network (HHNet) is proposed to learn semantic-rich high-resolution representations and preserve spatial details simultaneously. Specifically, a High-resolution Feature Learning (HFL) module with increasing dilation rates was first designed to learn the high-level high-resolution representations. Then, the HHNet was constructed by incorporating three HFL modules and two feature aggregation modules. The HHNet runs in a coarse-to-fine manner, and fine segmentation maps are output at the last level. Extensive experiments were conducted on fundus lesion segmentation, vessel segmentation, and optic cup segmentation. The experimental results reveal that the proposed method shows highly competitive or even superior performance in terms of segmentation performance and computation cost, indicating its potential advantages in clinical application.
引用
收藏
页数:14
相关论文
共 73 条
  • [51] Seoud Lama, 2016, IEEE Trans Med Imaging, V35, P1116, DOI 10.1109/TMI.2015.2509785
  • [52] Automated detection and classification of fundus diabetic retinopathy images using synergic deep learning model
    Shankar, K.
    Sait, Abdul Rahaman Wahab
    Gupta, Deepak
    Lakshmanaprabu, S. K.
    Khanna, Ashish
    Pandey, Hari Mohan
    [J]. PATTERN RECOGNITION LETTERS, 2020, 133 : 210 - 216
  • [53] Hard exudate segmentation in retinal image with attention mechanism
    Si, Ze
    Fu, Dongmei
    Liu, Yang
    Huang, Zhicheng
    [J]. IET IMAGE PROCESSING, 2021, 15 (03) : 587 - 597
  • [54] Simonyan K, 2015, Arxiv, DOI arXiv:1409.1556
  • [55] Srivastava N, 2014, J MACH LEARN RES, V15, P1929
  • [56] Ridge-based vessel segmentation in color images of the retina
    Staal, J
    Abràmoff, MD
    Niemeijer, M
    Viergever, MA
    van Ginneken, B
    [J]. IEEE TRANSACTIONS ON MEDICAL IMAGING, 2004, 23 (04) : 501 - 509
  • [57] Automated segmentation of exudates, haemorrhages, microaneurysms using single convolutional neural network
    Tan, Jen Hong
    Fujita, Hamido
    Sivaprasad, Sobha
    Bhandary, Sulatha V.
    Rao, A. Krishna
    Chua, Kuang Chua
    Acharya, U. Rajendra
    [J]. INFORMATION SCIENCES, 2017, 420 : 66 - 76
  • [58] Multi-scale channel importance sorting and spatial attention mechanism for retinal vessels segmentation
    Tang, Xianlun
    Zhong, Bing
    Peng, Jiangping
    Hao, Bohui
    Li, Jie
    [J]. APPLIED SOFT COMPUTING, 2020, 93 (93)
  • [59] PERI-Net: a parameter efficient residual inception network for medical image segmentation
    Uslu, Fatmatulzehra
    Bass, Cher
    Bharath, Anil A.
    [J]. TURKISH JOURNAL OF ELECTRICAL ENGINEERING AND COMPUTER SCIENCES, 2020, 28 (04) : 2261 - 2277
  • [60] Dual Encoding U-Net for Retinal Vessel Segmentation
    Wang, Bo
    Qiu, Shuang
    He, Huiguang
    [J]. MEDICAL IMAGE COMPUTING AND COMPUTER ASSISTED INTERVENTION - MICCAI 2019, PT I, 2019, 11764 : 84 - 92