Related Experiment Video
Updated: Jun 16, 2026
07:23
Retinal Vascular Reactivity as Assessed by Optical Coherence Tomography Angiography
Published on: March 26, 2020
Automated analysis of retinal vascular network connectivity
Bashir Al-Diri1, Andrew Hunter, David Steel
1Lincoln School of Computer Science, University of Lincoln, UK. baldiri@lincoln.ac.uk
Summary
This study presents a new algorithm for creating retinal vessel graphs by analyzing vessel connectivity. It uses self-organizing feature maps to accurately map retinal vasculature, even with overlapping vessels.
Area of Science:
- Ophthalmology
- Medical Imaging
- Computer Vision
Background:
- Accurate retinal vessel segmentation is crucial for diagnosing eye diseases.
- Existing methods struggle with complex vessel structures and overlapping vessels.
Purpose of the Study:
- To develop and evaluate an algorithm for generating a retinal vessel graph.
- To improve the analysis of retinal vasculature connectivity.
Main Methods:
- Utilized self-organizing feature maps (SOFMs) to model junction geometry cost functions.
- Developed specialized algorithms to address overlapping vessels.
- Tested the algorithm on the DRIVE database.
Main Results:
- Successfully formed retinal vessel graphs by analyzing segmented vessel connectivity.
- Demonstrated the algorithm's ability to resolve local segment end configurations.
- Handled overlapping vessels effectively.
Conclusions:
- The proposed algorithm accurately determines retinal network connectivity.
- This method offers a robust approach for retinal vessel graph formation.
- The algorithm shows promise for clinical applications in ophthalmology.
