Related Experiment Video
Updated: Aug 14, 2025
Author Spotlight: Advancing Human Cardiac Anatomy Through Multi-Scale Analysis of Hearts
Published on: June 28, 2024
A novel multi-attention, multi-scale 3D deep network for coronary artery segmentation
Caixia Dong1, Songhua Xu1, Duwei Dai1
1Institute of Medical Artificial Intelligence, the Second Affiliated Hospital of Xi'an Jiaotong University, Xi'an, 710004, China.
Insights
This study introduces CAS-Net, a novel 3D deep network for accurate coronary artery segmentation (CAS). CAS-Net significantly improves the diagnosis of coronary artery disease (CAD) by overcoming segmentation challenges like vessel variation and low contrast.
Area of Science:
- Medical imaging analysis
- Artificial intelligence in healthcare
- Cardiovascular disease research
Background:
- Accurate coronary artery segmentation (CAS) is crucial for diagnosing coronary artery disease (CAD).
- Existing CAS methods face challenges due to vessel variations, complex anatomy, and low contrast.
- Limited high-quality annotated data hinders the development of robust CAS models.
Purpose of the Study:
- To develop a novel 3D deep network, CAS-Net, for improved automatic coronary artery segmentation.
- To address the challenges of vessel variability, complex morphology, and low contrast in CAS.
- To enhance feature representation and fusion for more accurate vessel map generation.
Main Methods:
- Proposed CAS-Net, a multi-attention, multi-scale 3D deep network.
- Introduced attention-guided feature fusion (AGFF) and scale-aware feature enhancement (SAFE) modules.
- Developed a new dataset of 119 coronary computed tomographic angiography (CCTA) volumes for training and validation.
Main Results:
- CAS-Net demonstrated superior segmentation performance and generalization ability across multiple datasets.
- The proposed method significantly outperformed state-of-the-art algorithms, improving Dice similarity coefficient (DSC) by at least 4% compared to U-Net3D.
- The synergistic effect of AGFF, SAFE, and multi-scale feature aggregation (MSFA) modules contributed to enhanced segmentation accuracy.
Conclusions:
- CAS-Net offers a robust solution for automatic coronary artery segmentation, aiding in CAD diagnosis.
- The novel network architecture effectively handles variations and complexities in coronary artery imaging.
- The developed dataset and open-source implementation facilitate further research in cardiovascular image analysis.
Abstract:
Automatic segmentation of coronary arteries provides vital assistance to enable accurate and efficient diagnosis and evaluation of coronary artery disease (CAD). However, the task of coronary artery segmentation (CAS) remains highly challenging due to the large-scale variations exhibited by coronary arteries, their complicated anatomical structures and morphologies, as well as the low contrast between vessels and their background. To comprehensively tackle these challenges, we propose a novel multi-attention, multi-scale 3D deep network for CAS, which we call CAS-Net. Specifically, we first propose an attention-guided feature fusion (AGFF) module to efficiently fuse adjacent hierarchical features in the encoding and decoding stages to capture more effectively latent semantic information. Then, we propose a scale-aware feature enhancement (SAFE) module, aiming to dynamically adjust the receptive fields to extract more expressive features effectively, thereby enhancing the feature representation capability of the network. Furthermore, we employ the multi-scale feature aggregation (MSFA) module to learn a more distinctive semantic representation for refining the vessel maps. In addition, considering that the limited training data annotated with a quality golden standard are also a significant factor restricting the development of CAS, we construct a new dataset containing 119 cases consisting of coronary computed tomographic angiography (CCTA) volumes and annotated coronary arteries. Extensive experiments on our self-collected dataset and three publicly available datasets demonstrate that the proposed method has good segmentation performance and generalization ability, outperforming multiple state-of-the-art algorithms on various metrics. Compared with U-Net3D, the proposed method significantly improves the Dice similarity coefficient (DSC) by at least 4% on each dataset, due to the synergistic effect among the three core modules, AGFF, SAFE, and MSFA. Our implementation is released at https://github.com/Cassie-CV/CAS-Net.
