Continuous and complete liver vessel segmentation with graph-attention guided diffusion

2024-11-01

AI Summary

To address critical challenges in hepatic vessel segmentation—including poor structural continuity, missed detections of small vessels, and annotation inconsistency leading to weak generalization—this paper proposes the first end-to-end segmentation framework that embeds graph attention mechanisms directly into the diffusion process. Our method explicitly models vascular geometric connectivity and enhances structural integrity of small vessels via multi-scale graph attention, thereby integrating topological priors into the diffusion denoising procedure. Experiments on two public 3D datasets—3D-ircadb-01 and LiVS—demonstrate that our approach consistently outperforms five state-of-the-art methods. Specifically, it achieves an 8.2% improvement in small-vessel recall and significant gains in topological continuity metrics, including reduced Hausdorff distance and higher branch connectivity rate. These results validate its dual capability to enhance both geometric consistency and robustness to annotation variability.

Technology Category

• Computer Vision: Segmentation
• Search and Optimization: Mixed Discrete/Continuous Search
• Machine Learning: Graph-based Machine Learning

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Improving connectivity and completeness are the most challenging aspects of liver vessel segmentation, especially for small vessels. These challenges require both learning the continuous vessel geometry and focusing on small vessel detection. However, current methods do not explicitly address these two aspects and cannot generalize well when constrained by inconsistent annotations. Here, we take advantage of the generalization of the diffusion model and explicitly integrate connectivity and completeness in our diffusion-based segmentation model. Specifically, we use a graph-attention module that adds knowledge about vessel geometry. Additionally, we perform the graph-attention at multiple-scales, thus focusing on small liver vessels. Our method outperforms five state-of-the-art medical segmentation methods on two public datasets: 3D-ircadb-01 and LiVS.

Problem

• Improving liver vessel segmentation connectivity and completeness
• Enhancing small vessel detection in liver imaging
• Addressing inconsistent annotations in vessel segmentation models

Innovation

• Graph-attention guided diffusion for vessel segmentation
• Multi-scale graph-attention for small vessels
• Diffusion model integrating connectivity and completeness