Related Experiment Video
Updated: Oct 6, 2025
10:39
A Label-Free Segmentation Approach for Intravital Imaging of Mammary Tumor Microenvironment
Published on: May 24, 2022
2.5K
Microscopic retinal blood vessels detection and segmentation using support vector machine and K-nearest neighbors
Amjad Rehman1, Majid Harouni2, Mohsen Karimi3
1Artificial Intelligence & Data Analytics Lab CCIS, Prince Sultan University, Riyadh, Saudi Arabia.
Microscopy Research and Technique
|January 17, 2022
Summary
This study introduces an improved method for segmenting retinal blood vessels in fundus images, achieving 92% accuracy. This technique aids in diagnosing eye diseases like glaucoma and diabetic retinopathy.
Area of Science:
- Ophthalmology
- Medical Imaging
- Computer Vision
Background:
- Retinal blood vessel segmentation is crucial for diagnosing eye conditions such as glaucoma and diabetic retinopathy.
- Challenges in segmentation include the complex tree structure, variable vessel size, and random distribution of retinal vessels.
- The green band of retinal images is often utilized for enhanced microscopic vessel detection due to its high information content.
Purpose of the Study:
- To propose an improved supervised method for accurate retinal vessel segmentation.
- To enhance the diagnostic capabilities for retinal diseases through precise vessel mapping.
- To overcome the limitations of existing machine learning approaches in retinal vessel analysis.
Main Methods:
- Pre-processing and image quality enhancement of fundus images.
- Application of filtering and comparative histogram techniques for vessel segmentation.
- Extraction of statistical features including vessel tracking, maximum curvature, and curvelet coefficients.
- Classification using Support Vector Machine (SVM) and K-Nearest Neighbors (KNN) algorithms.
- Post-classification enhancement using morphological operators for improved segmentation accuracy.
Main Results:
- The proposed method achieved a high average accuracy of 92% in retinal vessel segmentation.
- The combination of feature extraction, machine learning classification, and morphological operations led to enhanced segmentation.
- The method demonstrated superior performance compared to existing retinal vessel segmentation strategies.
Conclusions:
- The developed supervised algorithm offers a robust and accurate solution for retinal vessel segmentation.
- The approach effectively addresses the complexities of retinal vasculature, improving diagnostic potential.
- The 92% accuracy signifies a significant advancement in automated analysis of fundus images for eye disease detection.
