MCP-Net: Real-Time Coronary Perception Network and Topological Analysis for Robotic Guidewire Navigation

Sahar Nasirihaghighi; Tianyuan Wang; Kenan Niu; Stefano Stramigioli

cs.CVOct 27, 2026

MCP-Net: Real-Time Coronary Perception Network and Topological Analysis for Robotic Guidewire Navigation

Authors: Sahar Nasirihaghighi, Tianyuan Wang, Kenan Niu, Stefano Stramigioli

Organizations: Department of Electrical Engineering, Mathematics and Computer Science (EEMCS), Robotics and Mechatronics group, University of Twente, 7500 AE Enschede, The Netherlands.

Abstract

Coronary guidewire navigation during percutaneous coronary intervention requires continuous interpretation of vessel anatomy, guidewire position, and target stenosis under two-dimensional X-ray fluoroscopy. While existing approaches commonly address vessel segmentation or device localization as separate tasks, a robotic system requires a more structured perception of the coronary scene to support downstream navigation. In this work, we present a simulation-driven framework centered on a Multi-Task Coronary Perception Network (MCP-Net) for robotic guidewire navigation, integrating coronary vessel segmentation, vascular topology extraction, and roadmap generation. Starting from three-dimensional coronary CT angiography, we generate simulated fluoroscopy images containing vascular structures, simulated guidewires, stenotic lesions with varying severity, and diverse vascular geometries, together with simulation-derived ground truth. MCP-Net jointly performs coronary vessel segmentation, guidewire-tip localization, and stenosis localization from a single fluoroscopic frame. These predictions are subsequently converted into a structured vascular graph containing centerlines, bifurcations, endpoints, branch directions, and branch angles. By associating the predicted stenosis with this graph, the framework generates a roadmap from the automatically selected graph root toward the target lesion. The results demonstrate that the proposed framework can jointly recover coronary perception targets and transform them into an interpretable vascular representation suitable for downstream robotic navigation.

Figures & tables

Fig. 1: Overview of the proposed framework. Coronary CT angiography is used to generate simulated fluoroscopy images and task-specific annotations for MCP-Net training. During real-time inference, the trained MCP-Net jointly predicts the vessel mask, guidewire-tip location, and stenosis landmarks from a single fluoroscopy frame. The predicted vasculature is converted into a structured centerline graph and combined with the predicted stenosis target to generate a navigation-oriented roadmap.

Task	Metric	Value
Vessel segmentation	Dice	$90.3\pm 3.0\%$
	clDice	$96.0\pm 3.0\%$
	IoU	$82.5\pm 4.8\%$
Guidewire tip ( $n=327$ )	Error	$1.09$ ( $0.73$ – $1.95$ ) px
Guidewire tip ( $n=327$ )	PCK@5	$88.1\%$
Stenosis throat ( $n=320$ )	Error	$3.82$ ( $1.88$ – $27.1$ ) px

TABLE I: Performance of the DeepLabV3+-based MCP-Net on the simulated test set (

n=336

). Localization errors are reported as median (IQR).

Fig. 2: Localization accuracy as a function of pixel tolerance for guidewire-tip and stenosis-throat localization on the simulated test set.

Network	Dice (%)	clDice (%)	IoU (%)	Tip error (px) $\downarrow$	Tip PCK@5 $\uparrow$	Throat error (px) $\downarrow$	Throat PCK@10 $\uparrow$	Throat in lesion (%) $\uparrow$	Error $>$ 15 px (%) $\downarrow$
DeepLabV3+	90.3	96.0	82.5	1.09	88.1	3.82	64.4	70.6	32.2
U-Net	87.7	94.4	78.3	1.79	82.3	5.92	62.2	71.5	34.7
SegFormer	86.5	93.1	76.5	2.06	74.3	6.37	57.5	68.0	35.0
FPN	85.7	93.4	75.2	1.54	79.5	8.97	53.1	66.5	41.6

TABLE II: Comparison of multi-task architectures on the simulated test set (

n=336

). Best results are shown in bold.

Fig. 3: Qualitative comparison of the multi-task architectures on representative simulated fluoroscopy images. For each test case, the input image and ground-truth vessel mask are followed by vessel-segmentation and guidewire-tip/stenosis localization predictions from DeepLabV3+, U-Net, SegFormer, and FPN.

Fig. 4: Representative guidewire-tip and stenosis geometric analysis. Predicted landmarks are associated with the vessel geometry to estimate guidewire-tip position relative to the vessel, local vessel width, stenosis throat and reference diameters, lesion extent, and descriptive image-based percentage diameter stenosis (%DS).

Fig. 5: Representative vascular topology extracted from MCP-Net vessel predictions. The predicted mask is skeletonized and converted into a graph containing the automatically selected root, bifurcations, endpoints, and connecting vessel segments. Graph edges are color-coded according to their vessel diameter.

Fig. 6: Real-time roadmap visualization for two representative cases. The root-to-stenosis path is overlaid on the simulated fluoroscopy image (left) and predicted vessel mask (right), together with the guidewire tip, stenosis, start and end points, and bifurcation-level navigation cues.

Medical Image Analysis

Medical image analysis uses computational methods to interpret clinical images, measure anatomy and disease, and support diagnosis and treatment.

See all 292 papers

Vision-Based Reasoning with Topology-Encoded Graphs for Anatomical Path Disambiguation in Robot-Assisted Endovascular Navigation
Feb 23, 2026Jiyuan Zhao, Zhengyu Shi, Wentong Tian +5Robotic Endovascular Intervention Coronary Angiography
Department of Control Science and Engineering, College of Electronics and Information Engineering, and Shanghai Institute of Intelligent Science and Technology, Tongji University, Shanghai 200092, China · Department of Electronic Engineering, Faculty of Engineering, The Chinese University of Hong Kong, Hong Kong SAR 999077, China · Shanghai Operation Robot Co., Ltd., Shanghai 201318, China +2
VDSB-GWSyn: Diffusion Schrödinger Bridge for Controllable and Anatomically Feasible Guidewire Synthesis in Coronary Angiography
May 27, 2026Haoyuan Tang, Zhuo Zhang, Jialin Li +2Medical Imaging Coronary Angiography
Tianjin University, Tianjin, China
ARIADNE: A Perception-Reasoning Synergy Framework for Trustworthy Coronary Angiography Analysis
Mar 19, 2026Zhan Jin, Yu Luo, Yizhou Zhang +5Medical Image Analysis Coronary Angiography
School of Mathematical Sciences, Ocean University of China, Qingdao, 266100, Shandong, China. · Department of Cardiology, Qilu Hospital (Qingdao), Cheeloo College of Medicine, Shandong University, No. 758 Hefei Road, Qingdao, 266000, Shandong, China.

Feb 23, 2026J/K move · Enter open · S save

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

Jiyuan Zhao, Zhengyu Shi, Wentong Tian, Tianliang Yao+4

Department of Control Science and Engineering, College of Electronics and Information Engineering, and Shanghai Institute of Intelligent Science and Technology, Tongji University, Shanghai 200092, China · Department of Electronic Engineering, Faculty of Engineering, The Chinese University of Hong Kong, Hong Kong SAR 999077, China · Shanghai Operation Robot Co., Ltd., Shanghai 201318, China+2

Robotic Endovascular Intervention Coronary Angiography Coronary Artery Segmentation

Open paper AskPDF arXiv

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.

Abstract

Medical Image Analysis

Explore similar work

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

VDSB-GWSyn: Diffusion Schrödinger Bridge for Controllable and Anatomically Feasible Guidewire Synthesis in Coronary Angiography

ARIADNE: A Perception-Reasoning Synergy Framework for Trustworthy Coronary Angiography Analysis