ScholarMate
English简体中文繁體中文
客服热线:400-1616-289
登录注册
期刊论文

An efficient multitasking cascade network for arteriovenous segmentation using dual-modal fundus images

Diwakar, Rajnish Kumar; Kumari, Pammi*; Saxena, Priyank*; Poddar, Raju
MULTIMEDIA TOOLS AND APPLICATIONS2023-11-2DOI: 10.1007/s11042-023-17343-0
7阅读
引用收藏
分享

摘要

Our eyesight does not remain the same throughout our lives. Certain diseases start affecting our vision with age, such as diabetic retinopathy, hypertension, and glaucoma. This has varied effects on the structure of the retinal blood vessels. Therefore, studying these retinal blood vessels is vital for pre-diagnosis and diagnosing ophthalmic disease. Low contrast, complicated vessel structure, and inhomogeneous background lighting across subjects pose complex challenges in classifying vessels from retinal images. This paper proposes a cascade-supervised miniature U-Net (CSMi-UNet) to perform arteriovenous segmentation on Dual-Modal fundus images. The Dual-Modal contains monochrome images acquired at two different wavelengths (570 nm and 610 nm) based on hemoglobin absorption spectra, other than the regular RGB fundus images. Because of the high similarity in the arteriole and venule (AV) morphological structure, single modal (RGB) fundus images often result in misclassification. The AV structures are more detailed in the monochromic images and contribute significantly to arteriovenous analysis. The proposed CSMi-UNet utilizes the information with considerably fewer parameters and attains state-of-the-art performance, making it suitable for portable devices. We achieved an arteriovenous classification accuracy of 98.36%, F1-scores of 82.46%, and 82.53% for AV, significantly outperforming previous studies on the Dual-Modal dataset.

学科与关键词

关键词
Arteriole and venule segmentationRetinal vesselsMonochrome imagesDual-Modal datasetCascade supervised miniature U-NetFundus images
相关
加载中…
  • Dual Attention Multiscale Network for Vessel Segmentation in Fundus Photography
    Yin, Pengshuai; Fang, Yupeng; Wan, Qilin*Mathematics, 2022, 10(19): 3687.
  • Retinal Blood Vessels and Optic Disc Segmentation Using U-Net
    David, S. Alex; Mahesh, C.; Kumar, V. Dhilip; Polat, Kemal*; Alhudhaif, Adi; Nour, MajidMATHEMATICAL PROBLEMS IN ENGINEERING, 2022, 2022: 8030954.被引 9
  • AVNet: A retinal artery/vein classification network with category-attention weighted fusion
    Kang, Hong; Gao, Yingqi; Guo, Song; Xu, Xia; Li, Tao; Wang, Kai*Computer Methods and Programs in Biomedicine, 2020, 195: 105629.被引 20
查看全部 →

发表信息

论文状态
公开发表
期刊
MULTIMEDIA TOOLS AND APPLICATIONS
发表日期
2023-11-02
DOI
10.1007/s11042-023-17343-0
语言
ENGLISH
成果收录
SCIE

作者

  • D

    Diwakar, Rajnish Kumar (第一作者)

  • K

    Kumari, Pammi (通讯)

查看全部 4 位作者收起作者
  • S

    Saxena, Priyank (通讯)

  • P

    Poddar, Raju

登录解锁更多功能

  • 收藏这篇,随时在“我的 · 收藏”找回
  • 认领你的成果,生成推广素材
  • 关注作者获取最新成果
  • 加入项目组、协同写作
注册 / 登录

产品服务

  • 科研之友
  • 创新城
  • 科创云

服务支持

  • 帮助中心
  • 隐私政策
  • 服务条款

联系方式

在线客服:【】
客服热线:400-1616-289
电子邮箱:support@scholarmate.com

关注或下载科研之友

微信二维码
微信公众号
客户端下载二维码
下载客户端
科研成果科研人员科研机构科研动态爱瑞思软件

©2026 深圳市科研之友网络服务有限公司

公安备案图标粤公网安备 44030502000213
粤ICP备 16046710 号粤B2-20110417