Related Experiment Video
Updated: Aug 29, 2025
Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
Published on: November 30, 2022
3D vessel-like structure segmentation in medical images by an edge-reinforced network
Likun Xia1, Hao Zhang2, Yufei Wu3
1College of Information Engineering, Capital Normal University, Beijing, China.
Insights
This study introduces a novel edge-reinforced neural network (ER-Net) for segmenting vessel-like structures in 3D medical images. ER-Net improves edge detection and segmentation accuracy, outperforming existing methods.
Area of Science:
- Medical imaging analysis
- Artificial intelligence in healthcare
- Biomedical engineering
Background:
- Vessel-like structures in medical images are crucial biomarkers for disease diagnosis and treatment.
- Current segmentation methods struggle with precise edge detection due to imbalanced voxel distribution.
- Accurate segmentation of these structures is vital for understanding disease mechanisms.
Purpose of the Study:
- To develop a generic neural network for segmenting vessel-like structures across various 3D medical imaging modalities.
- To address the challenge of segmenting crisp edges in 3D medical images.
- To improve the accuracy and robustness of vessel segmentation.
Main Methods:
- Proposed an edge-reinforced neural network (ER-Net) with an encoder-decoder architecture.
- Introduced a reverse edge attention module and an edge-reinforced optimization loss to enhance edge voxel weighting.
- Incorporated a feature selection module for adaptive selection of discriminative features.
Main Results:
- The ER-Net demonstrated superior performance in segmenting vessel-like structures compared to state-of-the-art algorithms.
- Experimental validation on four public datasets confirmed the method's effectiveness across different metrics.
- The proposed modules significantly improved segmentation performance by emphasizing edge voxels.
Conclusions:
- The developed ER-Net offers a robust and effective solution for segmenting vessel-like structures in 3D medical images.
- The novel attention and loss mechanisms successfully address the challenge of edge segmentation.
- This advancement holds promise for improved disease diagnosis and treatment planning in neurovascular and other pathologies.
Abstract:
The vessel-like structure in biomedical images, such as within cerebrovascular and nervous pathologies, is an essential biomarker in understanding diseases' mechanisms and in diagnosing and treating diseases. However, existing vessel-like structure segmentation methods often produce unsatisfactory results due to challenging segmentations for crisp edges. The edge and nonedge voxels of the vessel-like structure in three-dimensional (3D) medical images usually have a highly imbalanced distribution as most voxels are non-edge, making it challenging to find crisp edges. In this work, we propose a generic neural network for the segmentation of the vessel-like structures in different 3D medical imaging modalities. The new edge-reinforced neural network (ER-Net) is based on an encoder-decoder architecture. Moreover, a reverse edge attention module and an edge-reinforced optimization loss are proposed to increase the weight of the voxels on the edge of the given 3D volume to discover and better preserve the spatial edge information. A feature selection module is further introduced to select discriminative features adaptively from an encoder and decoder simultaneously, which aims to increase the weight of edge voxels, thus significantly improving the segmentation performance. The proposed method is thoroughly validated using four publicly accessible datasets, and the experimental results demonstrate that the proposed method generally outperforms other state-of-the-art algorithms for various metrics.
