Paper
17 February 2012 Robust tracking of a virtual electrode on a coronary sinus catheter for atrial fibrillation ablation procedures
Wen Wu, Terrence Chen, Norbert Strobel, Dorin Comaniciu
Author Affiliations +
Abstract
Catheter tracking in X-ray fluoroscopic images has become more important in interventional applications for atrial fibrillation (AF) ablation procedures. It provides real-time guidance for the physicians and can be used as reference for motion compensation applications. In this paper, we propose a novel approach to track a virtual electrode (VE), which is a non-existing electrode on the coronary sinus (CS) catheter at a more proximal location than any real electrodes. Successful tracking of the VE can provide more accurate motion information than tracking of real electrodes. To achieve VE tracking, we first model the CS catheter as a set of electrodes which are detected by our previously published learning-based approach.1 The tracked electrodes are then used to generate the hypotheses for tracking the VE. Model-based hypotheses are fused and evaluated by a Bayesian framework. Evaluation has been conducted on a database of clinical AF ablation data including challenging scenarios such as low signal-to-noise ratio (SNR), occlusion and nonrigid deformation. Our approach obtains 0.54mm median error and 90% of evaluated data have errors less than 1.67mm. The speed of our tracking algorithm reaches 6 frames-per-second on most data. Our study on motion compensation shows that using the VE as reference provides a good point to detect non-physiological catheter motion during the AF ablation procedures.2
© (2012) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Wen Wu, Terrence Chen, Norbert Strobel, and Dorin Comaniciu "Robust tracking of a virtual electrode on a coronary sinus catheter for atrial fibrillation ablation procedures", Proc. SPIE 8316, Medical Imaging 2012: Image-Guided Procedures, Robotic Interventions, and Modeling, 83162U (17 February 2012); https://doi.org/10.1117/12.911079
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Electrodes

Atrial fibrillation

Detection and tracking algorithms

Image segmentation

Automatic tracking

X-rays

Model-based design

Back to Top