Deep Learning Models for Retinal Blood Vessels Segmentation: A Review

被引:90
作者
Soomro, Toufique Ahmed [1 ,5 ]
Afifi, Ahmed J. [2 ]
Zheng, Lihong [1 ]
Soomro, Shafiullah [3 ]
Gao, Junbin [4 ]
Hellwich, Olaf [2 ]
Paul, Manoranjan [1 ]
机构
[1] Charles Sturt Univ, Sch Comp & Math, Bathurst, NSW 2795, Australia
[2] Tech Univ Berlin, Comp Vis & Remote Sensing, D-10587 Berlin, Germany
[3] Quaid E Awam Univ Engn & Sci Technol, Dept Basic Sci & Related Studies, Nawabshah 67480, Pakistan
[4] Univ Sydney, Business Sch, Camperdown, NSW 2006, Australia
[5] QUEST, Elect Engn Dept, Larkana Campus, Larkana, Pakistan
关键词
Retinal colour fundus images; convolutional neural networks; retinal vessels segmentation; NETWORKS; IMAGES;
D O I
10.1109/ACCESS.2019.2920616
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
This paper presents a comprehensive review of the principle and application of deep learning in retinal image analysis. Many eye diseases often lead to blindness in the absence of proper clinical diagnosis and medical treatment. For example, diabetic retinopathy (DR) is one such disease in which the retinal blood vessels of human eyes are damaged. The ophthalmologists diagnose DR based on their professional knowledge, that is labor intensive. With the advances in image processing and artificial intelligence, computer vision-based techniques have been applied rapidly and widely in the field of medical images analysis and are becoming a better way to advance ophthalmology in practice. Such approaches utilize accurate visual analysis to identify the abnormality of blood vessels with improved performance over manual procedures. More recently, machine learning, in particular, deep learning, has been successfully implemented in this area. In this paper, we focus on recent advances in deep learning methods for retinal image analysis. We review the related publications since 1982, which include more than 80 papers for retinal vessels detections in the research scope spanning from segmentation to classification. Although deep learning has been successfully implemented in other areas, we found only 17 papers so far focus on retinal blood vessel segmentation. This paper characterizes each deep learning based segmentation method as described in the literature. Analyzing along with the limitations and advantages of each method. In the end, we offer some recommendations for future improvement for retinal image analysis.
引用
收藏
页码:71696 / 71717
页数:22
相关论文
共 77 条
[1]  
Afifi A. J., 2016, 2016 INT C DIGITAL I, P1
[2]  
[Anonymous], 2015, RECENT ADV CONVOLUTI
[3]  
[Anonymous], 2018, P FED C COMP SCI INF
[4]  
[Anonymous], 2015, ICLR
[5]  
[Anonymous], 1989, P ADV NEURAL INFORM
[6]  
[Anonymous], PROC CVPR IEEE
[7]  
[Anonymous], 2015, ARXIV PREPRINT ARXIV
[8]  
[Anonymous], IEEE J BIOMED HLTH I
[9]  
[Anonymous], 2012, ARXIV
[10]  
[Anonymous], 2013, INT C MACHINE LEARNI