Related Experiment Videos
A fuzzy vessel tracking algorithm for retinal images based on fuzzy clustering
1Department of Electrical and Computer Engineering, Aristotle University of Thessaloniki, Greece.
IEEE Transactions on Medical Imaging
|August 4, 1998
Summary
This study introduces an unsupervised fuzzy algorithm for tracking ocular fundus vessels, improving detection accuracy. The method uses fuzzy C-means clustering for reliable vessel identification without manual initialization.
Area of Science:
- Ophthalmology
- Medical Imaging
- Computer Vision
Background:
- Ocular fundus vessel detection is crucial for diagnosing eye diseases.
- Existing vessel tracking methods face challenges with initialization and vessel profile modeling.
Purpose of the Study:
- To present a novel unsupervised fuzzy algorithm for ocular fundus vessel tracking.
- To overcome limitations of existing methods in initialization and vessel profile modeling.
Main Methods:
- Utilized an unsupervised fuzzy algorithm for vessel tracking.
- Employed fuzzy C-means clustering for vessel and non-vessel region determination.
- Incorporated procedures for validating detected vessels and handling junctions/forks.
Main Results:
- The algorithm automatically tracks fundus vessels using linguistic descriptions.
- Demonstrated very good overall performance on fundus images and simulated vessels.
- Achieved consistent estimation of vessel parameters.
Conclusions:
- The proposed fuzzy algorithm offers an effective solution for ocular fundus vessel detection.
- The method provides robust vessel tracking, addressing key challenges in the field.
- Results indicate high accuracy and reliability in vessel parameter estimation.