Related Experiment Video
Updated: Jun 10, 2026
10:39
A Label-Free Segmentation Approach for Intravital Imaging of Mammary Tumor Microenvironment
Published on: May 24, 2022
Unsupervised fuzzy based vessel segmentation in pathological digital fundus images
Giri Babu Kande1, P Venkata Subbaiah, T Satya Savithri
1Department of Electronics & Communication Engineering, Vasireddy Venkatadri Institute of Technology, Nambur, Guntur, A.P, India. kgiribabu@yahoo.com
Journal of Medical Systems
|August 13, 2010
Summary
This study introduces an automated method for segmenting retinal vasculature, even in pathological images. The novel approach improves accuracy for diagnosing eye conditions using advanced image processing techniques.
Area of Science:
- Medical Imaging
- Computer Vision
- Ophthalmology
Background:
- Retinal image segmentation is crucial for diagnosing eye diseases.
- Pathological conditions complicate automated vasculature segmentation.
- Existing methods struggle with non-uniform illumination and preserving vascular structures.
Purpose of the Study:
- To develop a novel, automated method for segmenting retinal vasculature.
- To address challenges in segmenting pathological retinal images.
- To improve the accuracy and robustness of retinal blood vessel segmentation.
Main Methods:
- Utilizes red and green channels for non-uniform illumination correction.
- Employs matched filtering to enhance blood vessel contrast.
- Applies spatially weighted fuzzy c-means clustering for segmentation, preserving vascular structures.
Main Results:
- Achieved an Area Under the Curve (AUC) of 0.9518 on the DRIVE database.
- Achieved an AUC of 0.9602 on the STARE database.
- Outperforms state-of-the-art unsupervised methods and is comparable to supervised methods.
Conclusions:
- The proposed method offers a robust and accurate solution for retinal vasculature segmentation.
- Effective for both normal and pathological retinal images.
- Demonstrates significant improvements over existing unsupervised techniques.