Reconstruction of Blood Vessel Networks from X-Ray Projections and a Vascular Catalogue

Abstract

Reconstruction of blood vessel networks from their x-ray projections is a challenging problem. This is because the correspondence problem must be solved for vessels which appear to twist, turn and overlap. Moreover, vessel networks vary in both branching structure and vessel shape. The extent of these variations is not catalogued in the clinical literature, or elsewhere. We have built a working system that accepts a few x-rays — separated by an angle of about ninety degrees — and reconstructs a three dimensional model of the vessel network. This task is impossible unless a priori information is used; how this information is represented is widely regarded as a key issue. Our representation makes a contribution by building and using a catalogue of anatomy that explicitly accounts for the wide variations in branching structure and shape. It is extensible in the sense that new information can be added to it at any time, and it is task independent in the sense that it can be used in many applications. We demonstrate its use in the problem of reconstructing vasculature from angiograms. Our reconstruction algorithm seeks to explain angiograms in terms of the vascular model. It can reconstruct vessel structures, even when vessels appear highly tangled, are missing, or extra vessels are present. The ability to recover complicated structure is the contribution made by our reconstruction method.

Cite

Text

Hall et al. "Reconstruction of Blood Vessel Networks from X-Ray Projections and a Vascular Catalogue." European Conference on Computer Vision, 1996. doi:10.1007/3-540-61123-1_148

Markdown

[Hall et al. "Reconstruction of Blood Vessel Networks from X-Ray Projections and a Vascular Catalogue." European Conference on Computer Vision, 1996.](https://mlanthology.org/eccv/1996/hall1996eccv-reconstruction/) doi:10.1007/3-540-61123-1_148

BibTeX

@inproceedings{hall1996eccv-reconstruction,
  title     = {{Reconstruction of Blood Vessel Networks from X-Ray Projections and a Vascular Catalogue}},
  author    = {Hall, Peter and Ngan, Milton and Andreae, Peter},
  booktitle = {European Conference on Computer Vision},
  year      = {1996},
  pages     = {293-303},
  doi       = {10.1007/3-540-61123-1_148},
  url       = {https://mlanthology.org/eccv/1996/hall1996eccv-reconstruction/}
}