Interventional Endocardial Motion Estimation from Electroanatomical Mapping Data: Application to Scar Characterization

Abstract

Scar presence and its characteristics play a fundamental role in several cardiac pathologies. To accurately define the extent and location of the scar is essential for a successful ventricular tachycardia ablation procedure. Nowadays, a set of widely accepted electrical voltage thresholds applied to local electrograms recorded are used intraoperatively to locate the scar. Information about cardiac mechanics could be considered to characterize tissues with different viability properties. We propose a novel method to estimate endocardial motion from data obtained with an electroanatomical mapping system together with the endocardial geometry segmented from preoperative 3-D magnetic resonance images, using a statistical atlas constructed with bilinear models. The method was validated using synthetic data generated from ultrasound images of nine volunteers and was then applied to seven ventricular tachycardia patients. Maximum bipolar voltages, commonly used to intraoperatively locate scar tissue, were compared to endocardial wall displacement and strain for all the patients. The results show that the proposed method allows endocardial motion and strain estimation and that areas with low-voltage electrograms also present low strain values.

A High-Resolution Atlas and Statistical Model of the Human Heart From Multislice CT

“…Figure 6 shows three phases of four subjects, together with bar plots for their subject and phase parameters. The applications of cardiac bilinear shape models have already been demonstrated in image segmentation [18] and motion analysis [84], [85]. In earlier works we have also demonstrated pipelines towards electrophysiological [86] and mechanical simulations [87] from surface-based cardiac segmentations.…”

Multi-source Motion Decoupling Ablation Catheter Guidance for Electrophysiology Procedures

“…Roujol et al [8] simplified the motion model by removing the two signals from fixed frequency bands after analysing the catheter tip position in the Fourier domain. Another recent approach fitted a model of motion to the catheter tip by instantiating a bilinear atlas of patient-specific shapes and deformations [9]. These methods, however, need a good spread of the recorded points on the endocardium in order for the instantiation to have a small error.…”

On the Detection of Myocadial Scar Based on ECG/VCG Analysis

“…However, in addition to its suitability only as a bedside test, CMR is a resource-demanding process, both in terms of cost and specialized personnel required for its operation. Additionally, other approaches for an accurate detection of scar are characterized as high interventional methods [6].…”