Abstract:Digital Subtraction Angiography (DSA) is one of the gold standards in vascular disease diagnosing. With the help of contrast agent, time-resolved 2D DSA images deliver comprehensive insights into blood flow information and can be utilized to reconstruct 3D vessel structures. Current commercial DSA systems typically demand hundreds of scanning views to perform reconstruction, resulting in substantial radiation exposure. However, sparse-view DSA reconstruction, aimed at reducing radiation dosage, is still underexplored in the research community. The dynamic blood flow and insufficient input of sparse-view DSA images present significant challenges to the 3D vessel reconstruction task. In this study, we propose to use a time-agnostic vessel probability field to solve this problem effectively. Our approach, termed as vessel probability guided attenuation learning, represents the DSA imaging as a complementary weighted combination of static and dynamic attenuation fields, with the weights derived from the vessel probability field. Functioning as a dynamic mask, vessel probability provides proper gradients for both static and dynamic fields adaptive to different scene types. This mechanism facilitates a self-supervised decomposition between static backgrounds and dynamic contrast agent flow, and significantly improves the reconstruction quality. Our model is trained by minimizing the disparity between synthesized projections and real captured DSA images. We further employ two training strategies to improve our reconstruction quality: (1) coarse-to-fine progressive training to achieve better geometry and (2) temporal perturbed rendering loss to enforce temporal consistency. Experimental results have demonstrated superior quality on both 3D vessel reconstruction and 2D view synthesis.

The Research Progress of Blood Vessel Image Reconstruction from Biplane Orthogonal Projections

Image Reconstruction of Arterial Cross Section from Biplane Projections Based on a Multiobjective Optimization Approach

Progress in High-Resolution Vascular Imaging and Quantification(Invited)

Three-dimensional Reconstruction of Intravascular Ultrasound Images

A New Approach to Reconstruction of Transverse Cross Sections of Arteriae from Biplane Orthogonal Projections [J]

Reconstruction of Vessel Axis Based on the Matching of Multi-Resolution Projections and Analysis of Corresponding Points

Iterative Cone-Beam Reconstruction Algorithm for Binary Vessel Based on Prior Information

3D Vessel Reconstruction in OCT-Angiography via Depth Map Estimation

Reconstruction of Vessel Axis Based on the Matching of Iterative Projection Algorithm and Analysis of Corresponding Points.

A Three-Dimensional Reconstruction Algorithm for Complex Background Image and Its Medical Applications

A Novel Method of Vessel Axis Reconstruction from the Rotary Projections Based on Furcation Model

Reconstruction of Rotary Dsa Vessel Axis Based on the Matching of Multiple Projections

Virtual Blood Vessels in Complex Background using Stereo X-ray Images

Study on 3D Reconstruction of Medical Image and Orthographic Views

3d Reconstruction Model Of Vessel Based On Object-Oriented Quantization

Shape Context and Projection Geometry Constrained Vasculature Matching for 3D Reconstruction of Coronary Artery

3D Reconstruction Method from Biplanar Radiography Using DLT Algorithm: Application to the Femur

Research on 3D Reconstruction Algorithm of Medical CT Image Based on Parallel Contour

A NEW METHOD FOR RECONSTRUCTING CROSS SECTIONS OF ARTERIES FROM X-RAY ANGIOGRAPHY IMAGE

3D Vessel Reconstruction from Sparse-View Dynamic DSA Images via Vessel Probability Guided Attenuation Learning

Reconstruction of blood flow velocity with deep learning information fusion from spectral ct projections and vessel geometry