Joint segmentation and classification of retinal arteries/veins from fundus images

作者：

Highlights：

• A fast deep-learning method that simultaneously segments and classifies vessels into arteries and veins is proposed.

• An efficient graph-based method is used to propagate the CNN's labeling through the vascular tree.

• Our method outperforms the leading previous works on a public dataset for A/V classification and is by far the fastest.

• The proposed global arterio-venous ratio (AVR) calculated on the whole fundus image using our automatic A/V segmentation method can better track vessel changes associated to diabetic retinopathy than the standard local AVR.

摘要

•A fast deep-learning method that simultaneously segments and classifies vessels into arteries and veins is proposed.•An efficient graph-based method is used to propagate the CNN's labeling through the vascular tree.•Our method outperforms the leading previous works on a public dataset for A/V classification and is by far the fastest.•The proposed global arterio-venous ratio (AVR) calculated on the whole fundus image using our automatic A/V segmentation method can better track vessel changes associated to diabetic retinopathy than the standard local AVR.

论文关键词：CNN,Artery and vein classification,Vessel segmentation,Fundus images,Retina

论文评审过程：Received 30 December 2017, Revised 9 August 2018, Accepted 17 February 2019, Available online 19 February 2019, Version of Record 27 February 2019.

论文官网地址：https://doi.org/10.1016/j.artmed.2019.02.004