Vision-Based Reasoning with Topology-Encoded Graphs for Anatomical Path Disambiguation in Robot-Assisted Endovascular Navigation

2026-02-23

AI Summary

This study addresses the ambiguity in vessel bifurcation paths during robot-assisted percutaneous coronary intervention caused by reliance solely on 2D digital subtraction angiography (DSA). To resolve this, the authors propose a two-stage framework, SCAR-UNet-GAT: first, a U-Net enhanced with a spatial coordinate attention mechanism performs high-precision vessel segmentation and centerline extraction; subsequently, a topology graph incorporating geometric features is constructed and refined via a graph attention network (GAT) under anatomical constraints to infer correct pathways. This approach uniquely integrates visual attention with topological graph encoding, effectively distinguishing true bifurcations from spurious crossings induced by projection artifacts. Evaluated on clinical DSA data, the method achieves a vessel segmentation Dice coefficient of 93.1%, with path disambiguation and target-reaching success rates of 95.0% and 90.0%, respectively, significantly outperforming conventional planning strategies.

Technology Category

Intelligent Robots: Motion and Path PlanningPlanning, Routing, and Scheduling: Activity and Plan RecognitionComputer Vision: Segmentation

Application Category

Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphsSearch and Retrieval-Augmented AI: Multilingual and cross-lingual Web searchResponsible Web: Human-perceived consequences of algorithmic deployment on the web

Abstract

Robotic-assisted percutaneous coronary intervention (PCI) is constrained by the inherent limitations of 2D Digital Subtraction Angiography (DSA). Unlike physicians, who can directly manipulate guidewires and integrate tactile feedback with their prior anatomical knowledge, teleoperated robotic systems must rely solely on 2D projections. This mode of operation, simultaneously lacking spatial context and tactile sensation, may give rise to projection-induced ambiguities at vascular bifurcations. To address this challenge, we propose a two-stage framework (SCAR-UNet-GAT) for real-time robotic path planning. In the first stage, SCAR-UNet, a spatial-coordinate-attention-regularized U-Net, is employed for accurate coronary vessel segmentation. The integration of multi-level attention mechanisms enhances the delineation of thin, tortuous vessels and improves robustness against imaging noise. From the resulting binary masks, vessel centerlines and bifurcation points are extracted, and geometric descriptors (e.g., branch diameter, intersection angles) are fused with local DSA patches to construct node features. In the second stage, a Graph Attention Network (GAT) reasons over the vessel graph to identify anatomically consistent and clinically feasible trajectories, effectively distinguishing true bifurcations from projection-induced false crossings. On a clinical DSA dataset, SCAR-UNet achieved a Dice coefficient of 93.1%. For path disambiguation, the proposed GAT-based method attained a success rate of 95.0% and a target-arrival success rate of 90.0%, substantially outperforming conventional shortest-path planning (60.0% and 55.0%) and heuristic-based planning (75.0% and 70.0%). Validation on a robotic platform further confirmed the practical feasibility and robustness of the proposed framework.

Problem

• robot-assisted endovascular navigation
• 2D DSA
• path disambiguation
• vascular bifurcations
• projection-induced ambiguity

Innovation

• SCAR-UNet
• Graph Attention Network
• topology-encoded graph
• anatomical path disambiguation
• robot-assisted endovascular navigation