A dual-branch 3D network with a vision graph encoder achieves leading Dice, Hausdorff distance, and precision results on the ASOCA and ImageCAS coronary CTA datasets.
It consists of 40 annotated Cardiac CTA images with an initial resolution of 512× 512× N , where N ranges from 168 to 334
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
eess.IV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
ViG3D-UNet: Volumetric Vascular Connectivity-Aware Segmentation via 3D Vision Graph Representation
A dual-branch 3D network with a vision graph encoder achieves leading Dice, Hausdorff distance, and precision results on the ASOCA and ImageCAS coronary CTA datasets.