A Multi-task Network to Detect Junctions in Retinal Vasculature

Junctions in the retinal vasculature are key points to be able to extract its topology, but they vary in appearance, depending on vessel density, width and branching/crossing angles. The complexity of junction patterns is usually accompanied by a scarcity of labels, which discourages the usage of very deep networks for their detection. We propose a multi-task network, generating labels for vessel interior, centerline, edges and junction patterns, to provide additional information to facilitate junction detection. After the initial detection of potential junctions in junction-selective probability maps, candidate locations are re-examined in centerline probability maps to verify if they connect at least 3 branches. The experiments on the DRIVE and IOSTAR showed that our method outperformed a recent study in which a popular deep network was trained as a classifier to find junctions. Moreover, the proposed approach is applicable to unseen datasets with the same degree of success, after training it only once.

• Publication year

  2018

• Venue

  International Conference on Medical Image Computing and Computer-Assisted Intervention

• Publication date

  2018-06-06

• Fields of study

  Medicine, Computer Science, Engineering

• Identifiers
  DOI 10.1007/978-3-030-00934-2_11arXiv 1806.03175
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Line Junction Detection Without Prior-Delineation of Curvilinear Structure in Biomedical Images

  2018cited by this paper
• Automatic Detection and Distinction of Retinal Vessel Bifurcations and Crossings in Colour Fundus Photography

  2017influential reference
• Recent Advancements in Retinal Vessel Segmentation

  2017cited by this paper
• Automatic detection of vascular bifurcations and crossings in retinal images using orientation scores

  2016influential reference
• A Cross-Modality Learning Approach for Vessel Segmentation in Retinal Images

  2016cited by this paper
• Deep Residual Learning for Image Recognition

  2015cited by this paper
• Biologically-Inspired Supervised Vasculature Segmentation in SLO Retinal Fundus Images

  2015cited by this paper
• AUTOMATIC DETECTION AND CLASSIFICATION OF RETINAL VASCULAR LANDMARKS

  2014cited by this paper
• Automatic detection of vascular bifurcations in segmented retinal images using trainable COSFIRE filters

  2013cited by this paper
• A Practical Guide to Training Restricted Boltzmann Machines

  2012cited by this paper
• Automatic detection and characterisation of retinal vessel tree bifurcations and crossovers in eye fundus images

  2011influential reference
• Retinal vessel segmentation using the 2-D Gabor wavelet and supervised classification

  2005cited by this paper
• Ridge-based vessel segmentation in color images of the retina

  2004cited by this paper
• Muliscale Vessel Enhancement Filtering

  1998cited by this paper

• Vessel-guided joint detection and classification of retinal vascular junctions

  2026cites this paper
• SAM-SPJunc: Self-Prompting for Junction Detection in Retinal Images via Radius-Based Representations

  2025cites this paper
• Vessel-guided and graph-based retinal vessel junction detection and classification

  2025cites this paper
• Intelligent Inspection Guidance of Urethral Endoscopy Based on SLAM with Blood Vessel Attentional Features

  2024cites this paper
• Automatic vessel crossing and bifurcation detection based on multi-attention network vessel segmentation and directed graph search

  2023cites this paper
• Robust Detection Model of Vascular Landmarks for Retinal Image Registration: A Two-Stage Convolutional Neural Network

  2022cites this paper
• Detection and classification of multi-scale retinal junctions using region-based CNN

  2021cites this paper
• Color fundus image registration using a learning-based domain-specific landmark detection methodology

  2021cites this paper
• clDice - a Topology-Preserving Loss Function for Tubular Structure Segmentation

  2020cites this paper
• Encoding anatomical tree priors for tubular structure extraction for medical images analysis

  2020influential citation
• clDice - a Novel Topology-Preserving Loss Function for Tubular Structure Segmentation

  2020cites this paper
• Retinal vascular junction detection and classification via deep neural networks

  2020cites this paper
• Machine Learning in Medical Imaging: 10th International Workshop, MLMI 2019, Held in Conjunction with MICCAI 2019, Shenzhen, China, October 13, 2019, Proceedings

  2019cites this paper
• Shape-Aware Complementary-Task Learning for Multi-Organ Segmentation

  2019cites this paper
• A Maximum Entropy Deep Reinforcement Learning Neural Tracker

  2019cites this paper
• An Inception Inspired Deep Network to Analyse Fundus Images

  2019cites this paper
• Deep multi-instance heatmap regression for the detection of retinal vessel crossings and bifurcations in eye fundus images

  2019cites this paper
• Retinal vascular analysis: Segmentation, tracing, and beyond

  2019cites this paper
• clDice - a Novel Connectivity-Preserving Loss Function for Vessel Segmentation

  2019cites this paper
• A Deep Learning Design for Improving Topology Coherence in Blood Vessel Segmentation

  2019cites this paper
• Retinal vessel junction detection using vessel guidance and heatmap regression

  year unknowncites this paper