Simultaneous segmentation and classification of the retinal arteries and veins from color fundus images
作者：
Highlights：
• A novel approach for the joint segmentation and classification of retinal vessels.
• The proposed novel loss decomposes the joint task into three segmentation problems.
• In the joint task, our approach outperforms the current state-of-the-art approach.
• Our method allows to detect vessel crossings, and handles them intuitively.
• The proposed approach outputs complete segmentation maps for both arteries and veins.
摘要
•A novel approach for the joint segmentation and classification of retinal vessels.•The proposed novel loss decomposes the joint task into three segmentation problems.•In the joint task, our approach outperforms the current state-of-the-art approach.•Our method allows to detect vessel crossings, and handles them intuitively.•The proposed approach outputs complete segmentation maps for both arteries and veins.
论文关键词：Medical imaging,Ophthalmology,Retina,Vessel segmentation,Artery and vein classification,Deep learning
论文评审过程：Received 11 January 2021, Revised 20 May 2021, Accepted 21 May 2021, Available online 29 May 2021, Version of Record 2 June 2021.
论文官网地址：https://doi.org/10.1016/j.artmed.2021.102116
