Related Experiment Video
Updated: May 22, 2026
09:17
Using Retinal Imaging to Study Dementia
Published on: November 6, 2017
Blood vessel segmentation methodologies in retinal images--a survey
M M Fraz1, P Remagnino, A Hoppe
1Digital Imaging Research Centre, Faculty of Science, Engineering and Computing, Kingston University London, London, United Kingdom. moazam.fraz@kingston.ac.uk
Computer Methods and Programs in Biomedicine
|April 25, 2012
Summary
This review categorizes retinal vessel segmentation algorithms for automatic disease screening. It analyzes techniques and performance metrics on public datasets, offering a framework for current research and future directions in retinal imaging analysis.
Area of Science:
- Ophthalmology
- Medical Imaging
- Computer Vision
Background:
- Automatic retinal disease screening relies on accurate retinal vessel segmentation.
- Fundus camera images are crucial for analyzing retinal vasculature.
Purpose of the Study:
- To review, analyze, and categorize retinal vessel extraction algorithms.
- To provide a framework for understanding existing research, current trends, and future directions in retinal segmentation.
Main Methods:
- Survey of various retinal vessel segmentation methodologies.
- Performance evaluation of algorithms using accuracy, sensitivity, specificity, and ROC curves.
- Analysis conducted on publicly available retinal image databases (DRIVE and STARE).
Main Results:
- Categorization of retinal vessel segmentation techniques.
- Comparative analysis of algorithm performance on benchmark datasets.
- Identification of key performance measures and their implications.
Conclusions:
- The study provides a comprehensive overview of retinal vessel segmentation algorithms.
- It highlights current trends, future research avenues, and open challenges in the field.
- Understanding these algorithms is vital for advancing automatic retinal disease screening systems.
