Related Experiment Video
Updated: Jul 12, 2025
04:48
Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
Published on: November 30, 2022
2.8K
DEAU-Net: Attention networks based on dual encoder for Medical Image Segmentation
Zhaojin Fu1, Jinjiang Li2, Zhen Hua2
1School of Information and Electronic Engineering, Shandong Technology and Business University, Yantai 264005, China; School of Computer Science and Technology, Shandong Technology and Business University, Yantai 264005, China.
Computers in Biology and Medicine
|October 20, 2023
Summary
This study introduces a novel dual-encoder feature attention network for medical image segmentation, improving both macro and micro feature extraction for enhanced accuracy.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Computer Vision
Background:
- U-Net derived networks are state-of-the-art for medical image segmentation.
- Existing models exhibit limitations in learning and extracting detailed features.
Purpose of the Study:
- To propose a novel feature attention network with a dual-encoder architecture.
- To enhance the extraction of both macro and micro features in medical images.
Main Methods:
- A dual-encoder approach for simultaneous macro and micro feature extraction.
- A feature attention fusion mechanism to integrate extracted features.
- A three-stage network design incorporating residual attention modules.
Main Results:
- The proposed network demonstrates superior performance in processing both macro and micro features.
- Experimental results show significant improvements in edge detail feature extraction.
- The DEAU-Net architecture achieved better results on two benchmark datasets.
Conclusions:
- The dual-encoder feature attention network effectively addresses limitations in current medical image segmentation models.
- The proposed architecture offers enhanced capabilities for detailed feature processing.
- This approach shows promise for advancing the accuracy and efficacy of medical image segmentation.
