Related Experiment Video
Updated: Jul 1, 2025
04:48
Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
Published on: July 5, 2024
399
MAG-Net : Multi-fusion network with grouped attention for retinal vessel segmentation
Yun Jiang1, Jie Chen1, Wei Yan1
1College of Computer Science and Engineering, Northwest Normal University, Lanzhou 730070, China.
Mathematical Biosciences and Engineering : MBE
|March 8, 2024
Summary
This study introduces MAG-Net, a novel deep learning model for retinal vessel segmentation. MAG-Net enhances accuracy by improving feature learning and reducing information loss, outperforming existing methods in ophthalmic diagnostics.
Area of Science:
- Ophthalmology
- Medical Imaging
- Computer Vision
Background:
- Retinal vessel segmentation is crucial for diagnosing eye diseases.
- Existing Convolutional Neural Networks (CNNs) face challenges like limited receptive fields and information loss during downsampling.
Purpose of the Study:
- To propose a new multi-fusion network with grouped attention (MAG-Net) to improve retinal vessel segmentation.
- To address the limitations of existing CNNs in capturing detailed vascular structures.
Main Methods:
- Developed MAG-Net featuring a hybrid convolutional fusion module for expanded receptive fields.
- Incorporated a grouped attention enhancement module to guide feature learning and preserve details via skip connections.
- Utilized a multi-scale feature fusion module to aggregate information across different scales, minimizing loss during upsampling.
Main Results:
- MAG-Net achieved high performance on DRIVE, CHASE, and STARE datasets.
- Segmentation accuracy reached up to 0.9773, specificity up to 0.9906, and Dice coefficients up to 0.8576.
- The proposed method demonstrated superior segmentation outcomes compared to existing techniques.
Conclusions:
- MAG-Net effectively overcomes limitations of traditional CNNs for retinal vessel segmentation.
- The network's architecture enhances feature learning and information preservation, leading to improved diagnostic accuracy.
- MAG-Net shows significant potential for clinical applications in ophthalmic disease diagnosis.
