Related Experiment Video
Updated: May 2, 2026
12:28
Quantification of Vascular Parameters in Whole Mount Retinas of Mice with Non-Proliferative and Proliferative Retinopathies
Published on: March 12, 2022
3.3K
Interactive retinal vessel extraction by integrating vessel tracing and graph search
Lu Wang1, Vinutha Kallem1, Mayank Bansal1
1SRI International, Princeton, NJ, USA.
Medical Image Computing and Computer-Assisted Intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention
|March 1, 2014
Summary
This study introduces an interactive method for blood vessel segmentation in low-quality retinal images. The approach uses curve fragments for efficient and accurate vessel extraction, requiring minimal user input.
Area of Science:
- Ophthalmology
- Medical Imaging
- Computer Vision
Background:
- Automatic blood vessel extraction from low-quality retinal images is challenging.
- Existing methods struggle with tortuous vessels and neighboring vessel interference.
Purpose of the Study:
- To develop an efficient interactive approach for near-perfect blood vessel segmentation.
- To overcome limitations of pixel-based and minimal-cost path techniques.
Main Methods:
- An interactive method using seed points and a cost function minimization.
- Graph construction based on curve fragments from vessel tracing, not individual pixels.
Main Results:
- Successfully overcomes shortcut problems in tortuous vessel extraction.
- Reduces issues with neighboring vessel interference.
- Achieves less user interaction for thin and tortuous vessels in low contrast images.
- Demonstrates significantly faster processing times.
Conclusions:
- The proposed interactive approach offers efficient and accurate blood vessel segmentation.
- It is particularly effective for challenging low-quality retinal images.
- The method improves upon existing techniques by using curve fragments for graph construction.
