Related Experiment Video
Updated: Dec 28, 2025
Quantification of Vascular Parameters in Whole Mount Retinas of Mice with Non-Proliferative and Proliferative Retinopathies
Published on: March 12, 2022
BSCN: bidirectional symmetric cascade network for retinal vessel segmentation
Yanfei Guo1, Yanjun Peng2,3
1College of Information Science and Engineering,Shandong University of Science and Technology, Shandong, Qingdao 266590, China.
Insights
This study introduces a novel Bidirectional Symmetric Cascade Network (BSCN) for accurate retinal blood vessel segmentation, crucial for diagnosing cardiovascular diseases. The new method enhances detection of small vessels and improves diagnostic accuracy.
Area of Science:
- Medical Imaging
- Computer Vision
- Ophthalmology
Background:
- Retinal blood vessel segmentation is vital for diagnosing cardiovascular diseases like hypertension and diabetes.
- Manual segmentation is time-consuming, labor-intensive, and lacks accuracy.
- Automated methods are needed for precise and efficient retinal vessel analysis.
Purpose of the Study:
- To develop an automated, accurate method for retinal blood vessel segmentation.
- To enable fine segmentation of vessels across various diameters.
- To improve diagnostic capabilities for retinal conditions and systemic diseases.
Main Methods:
- Proposed a Bidirectional Symmetric Cascade Network (BSCN).
- Employed scale-specific supervision for different network layers.
- Introduced a Dense Dilated Convolution Module (DDCM) for multi-scale feature extraction.
- Fused outputs from DDCM for final segmentation.
Main Results:
- Achieved high accuracy (0.9846-0.9889) and AUC (0.9874-0.9941) on four datasets (DRIVE, STARE, HRF, CHASE_DB1).
- Demonstrated robustness against lesion background interference.
- Accurately detected tiny blood vessels at intersections.
Conclusions:
- The proposed BSCN method offers a robust and accurate solution for retinal blood vessel segmentation.
- Outperforms state-of-the-art methods in detecting fine and intersecting vessels.
- Significantly advances computer-aided diagnosis for retinal and cardiovascular diseases.
Background:
Retinal blood vessel segmentation has an important guiding significance for the analysis and diagnosis of cardiovascular diseases such as hypertension and diabetes. But the traditional manual method of retinal blood vessel segmentation is not only time-consuming and laborious but also cannot guarantee the accuracy and efficiency of diagnosis. Therefore, it is especially significant to create a computer-aided method of automatic and accurate retinal vessel segmentation.
Methods:
In order to extract the blood vessels' contours of different diameters to realize fine segmentation of retinal vessels, we propose a Bidirectional Symmetric Cascade Network (BSCN) where each layer is supervised by vessel contour labels of specific diameter scale instead of using one general ground truth to train different network layers. In addition, to increase the multi-scale feature representation of retinal blood vessels, we propose the Dense Dilated Convolution Module (DDCM), which extracts retinal vessel features of different diameters by adjusting the dilation rate in the dilated convolution branches and generates two blood vessel contour prediction results by two directions respectively. All dense dilated convolution module outputs are fused to obtain the final vessel segmentation results.
Results:
We experimented the three datasets of DRIVE, STARE, HRF and CHASE_DB1, and the proposed method reaches accuracy of 0.9846/0.9872/0.9856/0.9889 and AUC of 0.9874/0.9941/0.9882/0.9874 on DRIVE, STARE, HRF and CHASE_DB1.
Conclusions:
The experimental results show that compared with the state-of-art methods, the proposed method has strong robustness, it not only avoids the adverse interference of the lesion background but also detects the tiny blood vessels at the intersection accurately.
