Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Uncertainty-Aware Concept and Motion Segmentation for Semi-Supervised Angiography Videos

Yu Luo, Guangyu Wei, Yangfan Li, Jieyu He, Yueming Lyu

TL;DR

This work proposes SAM3-based Teacher-student framework with Motion-Aware consistency and Progressive Confidence Regularization (SMART), a semi-supervised vessel segmentation approach for X-ray angiography videos that achieves state-of-the-art performance while requiring significantly fewer annotations.

Abstract

Segmentation of the main coronary artery from X-ray coronary angiography (XCA) sequences is crucial for the diagnosis of coronary artery diseases. However, this task is challenging due to issues such as blurred boundaries, inconsistent radiation contrast, complex motion patterns, and a lack of annotated images for training. Although Semi-Supervised Learning (SSL) can alleviate the annotation burden, conventional methods struggle with complicated temporal dynamics and unreliable uncertainty quantification. To address these challenges, we propose SAM3-based Teacher-student framework with Motion-Aware consistency and Progressive Confidence Regularization (SMART), a semi-supervised vessel segmentation approach for X-ray angiography videos. First, our method utilizes SAM3's unique promptable concept segmentation design and innovates a SAM3-based teacher-student framework to maximize the performance potential of both the teacher and the student. Second, we enhance segmentation by integrating the vessel mask warping technique and motion consistency loss to model complex vessel dynamics. To address the issue of unreliable teacher predictions caused by blurred boundaries and minimal contrast, we further propose a progressive confidence-aware consistency regularization to mitigate the risk of unreliable outputs. Extensive experiments on three datasets of XCA sequences from different institutions demonstrate that SMART achieves state-of-the-art performance while requiring significantly fewer annotations, making it particularly valuable for real-world clinical applications where labeled data is scarce. Our code is available at: https://github.com/qimingfan10/SMART.

Uncertainty-Aware Concept and Motion Segmentation for Semi-Supervised Angiography Videos

TL;DR

This work proposes SAM3-based Teacher-student framework with Motion-Aware consistency and Progressive Confidence Regularization (SMART), a semi-supervised vessel segmentation approach for X-ray angiography videos that achieves state-of-the-art performance while requiring significantly fewer annotations.

Abstract

Segmentation of the main coronary artery from X-ray coronary angiography (XCA) sequences is crucial for the diagnosis of coronary artery diseases. However, this task is challenging due to issues such as blurred boundaries, inconsistent radiation contrast, complex motion patterns, and a lack of annotated images for training. Although Semi-Supervised Learning (SSL) can alleviate the annotation burden, conventional methods struggle with complicated temporal dynamics and unreliable uncertainty quantification. To address these challenges, we propose SAM3-based Teacher-student framework with Motion-Aware consistency and Progressive Confidence Regularization (SMART), a semi-supervised vessel segmentation approach for X-ray angiography videos. First, our method utilizes SAM3's unique promptable concept segmentation design and innovates a SAM3-based teacher-student framework to maximize the performance potential of both the teacher and the student. Second, we enhance segmentation by integrating the vessel mask warping technique and motion consistency loss to model complex vessel dynamics. To address the issue of unreliable teacher predictions caused by blurred boundaries and minimal contrast, we further propose a progressive confidence-aware consistency regularization to mitigate the risk of unreliable outputs. Extensive experiments on three datasets of XCA sequences from different institutions demonstrate that SMART achieves state-of-the-art performance while requiring significantly fewer annotations, making it particularly valuable for real-world clinical applications where labeled data is scarce. Our code is available at: https://github.com/qimingfan10/SMART.
Paper Structure (12 sections, 6 equations, 3 figures, 3 tables)

This paper contains 12 sections, 6 equations, 3 figures, 3 tables.

Table of Contents

  1. Introduction
  2. Method
  3. Text-driven Segmentation Fine-tuning
  4. Confidence-aware Consistency Regularization
  5. Dual-Stream Temporal Consistency
  6. Experiments
  7. Experimental Setup
  8. Datasets.
  9. Implementation Details and Evaluation Metrics
  10. Comparisons with the State-of-the-arts.
  11. Ablation Studies
  12. Conclusion

Figures (3)

  • Figure 1: The overview of our proposed method. SMART (SAM3-based Motion-Aware Confidence Regularization for Teacher-Student Architecture) model enhances the reliability of teacher model outputs through a confidence-aware consistency loss $\mathcal{L}_{\text{conf}}$. It incorporates temporal understanding from video sequences via dual-stream temporal consistency losses $\mathcal{L}_{\text{opti}}$ and $\mathcal{L}_{\text{coh}}$, while effectively leveraging annotated data through the supervised loss $\mathcal{L}_{\text{sup}}$.
  • Figure 2: Visualization results on the vessel segmentation.
  • Figure 3: Visual comparisons of ablation studies.