Related Experiment Video
Updated: Dec 22, 2025
06:57
Author Spotlight: Advancing Cardiovascular Imaging - Introducing the Spatially Weighted Calcium Score for Early Disease Detection
Published on: September 22, 2023
1.4K
A coronary artery segmentation method based on region growing with variable sector search area.
Guangkun Ma1,2, Jinzhu Yang1,3, Hong Zhao1
1School of Computer Science and Engineering, Northeastern University, Shenyang, Liaoning, China.
Summary
This study introduces an improved region growing method for coronary artery segmentation. The new approach enhances accuracy in detecting vessel branches and handles complex vessel shapes effectively.
Area of Science:
- Medical Imaging
- Computational Anatomy
Background:
- Coronary artery segmentation is crucial for diagnosing coronary artery disease.
- Traditional region growing methods are limited by noise sensitivity and manual intervention.
- Existing advanced methods struggle with over- or under-segmentation.
Purpose of the Study:
- To develop a more robust and accurate coronary artery segmentation technique.
- To overcome limitations of classical and recent region growing algorithms.
- To improve the detection of small branches and handle complex vessel geometries.
Main Methods:
- A novel region growing algorithm incorporating a variable sector search area.
- A growing rule combining Hessian vectors with the variable sector search area.
- Optimization technique to remove small, disconnected segmented regions.
Main Results:
- The proposed method successfully segments more vessel branches, including small ones.
- Effective performance is maintained even with vessel stenosis and significant curvature.
- Improved segmentation quality compared to existing region growing techniques.
Conclusions:
- The new method offers enhanced accuracy in coronary artery segmentation.
- Quantitative evaluations demonstrate superior performance metrics.
- This technique provides a more reliable tool for coronary artery disease diagnosis.
