Related Experiment Videos
Artery-vein separation via MRA--an image processing approach.
1Department of Radiology, University of Pennsylvania, Philadelphia 19104, USA.
IEEE Transactions on Medical Imaging
|August 22, 2001
Summary
This study introduces a fast, near-automatic method for separating arteries and veins in contrast-enhanced magnetic resonance angiographic (CE-MRA) images. The fuzzy connectedness algorithm accurately visualizes vascular structures, improving upon manual segmentation.
Area of Science:
- Medical Imaging
- Image Processing
- Computational Anatomy
Background:
- Accurate segmentation of vascular structures is crucial for diagnosing and treating cardiovascular diseases.
- Existing methods for artery-vein separation in CE-MRA data are often time-consuming and lack detail.
Purpose of the Study:
- To develop and validate a near-automatic, efficient process for separating arteries and veins in CE-MRA images.
- To enable optimal 3D visualization of vascular structures for clinical use.
Main Methods:
- Utilized fuzzy connected object delineation principles for segmentation.
- Employed absolute fuzzy connectedness for initial vessel segmentation from background.
- Applied iterative relative fuzzy connectedness for artery-vein separation, with competing voxel membership based on connectedness strength.
Main Results:
- Successfully segmented arteries and veins in 133 pelvic and 2 carotid CE-MRA studies with unified parameters.
- Demonstrated superior separation of higher-order branches compared to manual segmentation, yielding more detailed vascular structures.
- Achieved an average processing time of approximately 4.5 minutes per study.
Conclusions:
- The developed fuzzy connectedness approach provides a routine, accurate, and efficient method for artery-vein separation in CE-MRA.
- This technique offers a significant advancement over manual segmentation, providing greater detail and speed for clinical applications.