Related Experiment Video
Updated: Nov 20, 2025
Quantification of Vascular Parameters in Whole Mount Retinas of Mice with Non-Proliferative and Proliferative Retinopathies
Published on: March 12, 2022
Efficient BFCN for Automatic Retinal Vessel Segmentation
Yun Jiang1, Falin Wang1, Jing Gao1
1College of Computer Science and Engineering, Northwest Normal University, Lanzhou, Gansu, China.
Insights
This study introduces a novel butterfly fully convolutional neural network (BFCN) for enhanced retinal vessel segmentation. The BFCN method improves accuracy in diagnosing diseases like diabetic retinopathy and hypertension.
Area of Science:
- Medical Imaging
- Computer Vision
- Ophthalmology
Background:
- Retinal vessel segmentation is crucial for diagnosing diabetic retinopathy, hypertension, and cardiovascular diseases.
- Current deep convolutional neural network (DCNN) methods struggle with limited receptive fields and spatial information, hindering global context capture and lesion identification.
- This leads to poor segmentation efficiency and difficulty in identifying critical areas.
Purpose of the Study:
- To develop an improved method for retinal vessel segmentation.
- To enhance the accuracy and efficiency of identifying retinal vasculature for disease diagnosis.
- To address limitations in existing DCNN-based segmentation approaches.
Main Methods:
- Proposed a butterfly fully convolutional neural network (BFCN) architecture.
- Employed automatic color enhancement (ACE) to improve low contrast in retinal images.
- Integrated a multiscale information extraction (MSIE) module and a transfer layer (T_Layer) to capture global context and preserve spatial information.
- Utilized Laplacian sharpening for postprocessing segmentation images.
Main Results:
- Achieved high accuracy on benchmark datasets: 0.9627 (DRIVE), 0.9735 (STARE), and 0.9688 (CHASE).
- The BFCN method demonstrated improved capability in capturing global context and detailed vessel structures.
- ACE and Laplacian sharpening effectively addressed low contrast and enhanced segmentation precision.
Conclusions:
- The proposed BFCN method significantly improves retinal vessel segmentation accuracy.
- This approach offers a more effective tool for the early diagnosis of various eye and systemic diseases.
- The integration of ACE, MSIE, T_Layer, and Laplacian sharpening provides a robust solution for retinal image analysis.
Abstract:
Retinal vessel segmentation has high value for the research on the diagnosis of diabetic retinopathy, hypertension, and cardiovascular and cerebrovascular diseases. Most methods based on deep convolutional neural networks (DCNN) do not have large receptive fields or rich spatial information and cannot capture global context information of the larger areas. Therefore, it is difficult to identify the lesion area, and the segmentation efficiency is poor. This paper presents a butterfly fully convolutional neural network (BFCN). First, in view of the low contrast between blood vessels and the background in retinal blood vessel images, this paper uses automatic color enhancement (ACE) technology to increase the contrast between blood vessels and the background. Second, using the multiscale information extraction (MSIE) module in the backbone network can capture the global contextual information in a larger area to reduce the loss of feature information. At the same time, using the transfer layer (T_Layer) can not only alleviate gradient vanishing problem and repair the information loss in the downsampling process but also obtain rich spatial information. Finally, for the first time in the paper, the segmentation image is postprocessed, and the Laplacian sharpening method is used to improve the accuracy of vessel segmentation. The method mentioned in this paper has been verified by the DRIVE, STARE, and CHASE datasets, with the accuracy of 0.9627, 0.9735, and 0.9688, respectively.
