VasTSD: Learning 3D Vascular Tree-state Space Diffusion Model for Angiography Synthesis

2025-03-17

AI Summary

This study addresses the challenge of reconstructing 3D vascular structures from non-contrast imaging modalities (e.g., non-contrast CT/MR), thereby avoiding risks associated with contrast agents and ionizing radiation. To overcome limitations of existing methods—particularly poor preservation of vascular continuity and weak cross-modal generalization—we propose: (1) a novel vascular-tree state-space serialization framework that explicitly models the topological evolution of 3D vasculature; (2) a cross-modality-consistent vascular state representation, mitigating structural discontinuities inherent in conventional 2D slice-wise modeling; and (3) an integrated architecture combining 3D diffusion modeling, graph-structured sequence encoding, and multimodal feature alignment. Evaluated on multi-center, multi-modal (CTA/MRA), and multi-anatomic (intracranial/coronary/peripheral) datasets, our method achieves superior vascular continuity and anatomical fidelity over state-of-the-art approaches, with comprehensive quantitative improvements across all metrics.

Technology Category

• Computer Vision: Multi-modal Vision
• Machine Learning: Multimodal Learning
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Angiography imaging is a medical imaging technique that enhances the visibility of blood vessels within the body by using contrast agents. Angiographic images can effectively assist in the diagnosis of vascular diseases. However, contrast agents may bring extra radiation exposure which is harmful to patients with health risks. To mitigate these concerns, in this paper, we aim to automatically generate angiography from non-angiographic inputs, by leveraging and enhancing the inherent physical properties of vascular structures. Previous methods relying on 2D slice-based angiography synthesis struggle with maintaining continuity in 3D vascular structures and exhibit limited effectiveness across different imaging modalities. We propose VasTSD, a 3D vascular tree-state space diffusion model to synthesize angiography from 3D non-angiographic volumes, with a novel state space serialization approach that dynamically constructs vascular tree topologies, integrating these with a diffusion-based generative model to ensure the generation of anatomically continuous vasculature in 3D volumes. A pre-trained vision embedder is employed to construct vascular state space representations, enabling consistent modeling of vascular structures across multiple modalities. Extensive experiments on various angiographic datasets demonstrate the superiority of VasTSD over prior works, achieving enhanced continuity of blood vessels in synthesized angiographic synthesis for multiple modalities and anatomical regions.

Problem

• Automatically generate angiography from non-angiographic inputs
• Maintain 3D vascular structure continuity in angiography synthesis
• Enhance effectiveness across different imaging modalities

Innovation

• 3D vascular tree-state space diffusion model
• Dynamic vascular tree topology construction
• Multi-modality consistent vascular modeling