Related Experiment Video
Updated: Apr 26, 2026
09:57
Development and Evaluation of 3D-Printed Cardiovascular Phantoms for Interventional Planning and Training
Published on: January 18, 2021
3.8K
3D vessel reconstruction from sparse-view dynamic DSA images via vessel probability guided attenuation learning
Zhentao Liu1, Huangxuan Zhao2, Wenhui Qin1
1School of Biomedical Engineering & State Key Laboratory of Advanced Medical Materials and Devices, ShanghaiTech University, Shanghai, China.
Medical Image Analysis
|April 24, 2026
Summary
This study introduces a new method for Digital Subtraction Angiography (DSA) using neural rendering to reconstruct 3D blood vessels from fewer images. This approach significantly reduces radiation exposure while maintaining high-quality diagnostic imaging.
Area of Science:
- Medical Imaging
- Radiology
- Computer Vision
Background:
- Digital Subtraction Angiography (DSA) is crucial for diagnosing vascular diseases.
- Current DSA requires numerous scans, leading to high radiation doses.
- Reconstructing 3D vessel structures from 2D DSA images is vital for assessment.
Purpose of the Study:
- To develop a novel framework for high-quality sparse-view DSA reconstruction.
- To reduce patient radiation exposure by minimizing scanning views.
- To improve 3D vessel structure and 2D DSA image synthesis.
Main Methods:
- Proposed a neural rendering-based optimization framework: vessel probability guided attenuation learning.
- Modeled DSA imaging as a weighted combination of static and dynamic attenuation fields.
- Utilized vessel probability as a foreground mask for self-supervised decomposition and employed progressive training and temporal perturbed rendering loss.
Main Results:
- Achieved high-quality 3D vessel reconstruction from sparse-view DSA.
- Demonstrated effective self-supervised decomposition of static backgrounds and dynamic contrast flow.
- Successfully synthesized high-quality 2D DSA images.
Conclusions:
- The proposed framework significantly enhances sparse-view DSA reconstruction quality.
- The method effectively reduces radiation dosage while preserving diagnostic information.
- This approach offers a promising advancement in vascular imaging and diagnosis.
