Related Experiment Video
Updated: Jan 4, 2026
Quantification of Vascular Parameters in Whole Mount Retinas of Mice with Non-Proliferative and Proliferative Retinopathies
Published on: March 12, 2022
Dense Dilated Network With Probability Regularized Walk for Vessel Detection
Insights
This study introduces a novel retinal vessel detection method that improves connectivity by combining a dense dilated network and a probability regularized walk algorithm. The new approach enhances accuracy and diagnostic value for ocular diseases.
Area of Science:
- Ophthalmology
- Medical Imaging
- Computer Vision
Background:
- Retinal vessel detection is crucial for diagnosing ocular diseases.
- Existing methods often fail to preserve vessel connectivity, impacting diagnostic accuracy.
- Connectivity is vital for comprehensive retinal analysis.
Purpose of the Study:
- To propose a novel method for accurate retinal vessel detection.
- To address the issue of vessel discontinuity in current detection algorithms.
- To enhance the diagnostic utility of retinal imaging through improved vessel segmentation.
Main Methods:
- A dense dilated network with feature extraction blocks in an encoder-decoder structure was utilized for initial vessel detection.
- A multi-scale Dice loss function was employed for network training.
- A probability regularized walk algorithm was introduced to ensure vessel connectivity.
Main Results:
- The proposed method demonstrated superior performance across three public datasets (DRIVE, STARE, CHASE_DB1).
- Significant improvements were observed in accuracy, sensitivity, and specificity compared to state-of-the-art methods.
- Enhanced vessel connectivity was achieved, addressing segmentation fractures.
Conclusions:
- The novel method effectively detects retinal vessels with improved connectivity.
- This approach offers a significant advancement for the diagnosis and treatment of ocular diseases.
- The proposed technique provides more reliable and comprehensive retinal vessel segmentation.
Abstract:
The detection of retinal vessel is of great importance in the diagnosis and treatment of many ocular diseases. Many methods have been proposed for vessel detection. However, most of the algorithms neglect the connectivity of the vessels, which plays an important role in the diagnosis. In this paper, we propose a novel method for retinal vessel detection. The proposed method includes a dense dilated network to get an initial detection of the vessels and a probability regularized walk algorithm to address the fracture issue in the initial detection. The dense dilated network integrates newly proposed dense dilated feature extraction blocks into an encoder-decoder structure to extract and accumulate features at different scales. A multi-scale Dice loss function is adopted to train the network. To improve the connectivity of the segmented vessels, we also introduce a probability regularized walk algorithm to connect the broken vessels. The proposed method has been applied on three public data sets: DRIVE, STARE and CHASE_DB1. The results show that the proposed method outperforms the state-of-the-art methods in accuracy, sensitivity, specificity and also area under receiver operating characteristic curve.
