DeNVeR: Deformable Neural Vessel Representations for Unsupervised Video Vessel Segmentation

TL;DR

DeNVeR is an unsupervised neural approach for vessel segmentation in X-ray angiography videos that leverages optical flow, layer separation, and test-time training to improve accuracy and generalization, introducing a new annotated dataset.

Contribution

The paper introduces DeNVeR, a novel unsupervised method for vessel segmentation that incorporates layer separation, motion modeling, and a new dataset, advancing the state-of-the-art in accuracy and robustness.

Findings

DeNVeR outperforms existing methods in accuracy.

The approach maintains temporal coherence in segmentation.

The new XACV dataset provides high-quality ground truth for evaluation.

Abstract

This paper presents Deformable Neural Vessel Representations (DeNVeR), an unsupervised approach for vessel segmentation in X-ray angiography videos without annotated ground truth. DeNVeR utilizes optical flow and layer separation techniques, enhancing segmentation accuracy and adaptability through test-time training. Key contributions include a novel layer separation bootstrapping technique, a parallel vessel motion loss, and the integration of Eulerian motion fields for modeling complex vessel dynamics. A significant component of this research is the introduction of the XACV dataset, the first X-ray angiography coronary video dataset with high-quality, manually labeled segmentation ground truth. Extensive evaluations on both XACV and CADICA datasets demonstrate that DeNVeR outperforms current state-of-the-art methods in vessel segmentation accuracy and generalization capability while maintaining temporal coherency.