The Impact of Torso Signal Processing on Noninvasive Electrocardiographic Imaging Reconstructions

Laura R. Bear,Wilson Good,Eelco van Dam,Jaume Coll-Font,Jana Svehlikova,Yesim Serinagaoglu Dogrusoz,Rob MacLeod

doi:10.1109/tbme.2020.3003465

Abstract

Goal:To evaluate state-of-the-art signal processing methods for epicardial potential-based noninvasive electrocardiographic imaging reconstructions of single-site pacing data.Methods:Experimental data were obtained from two torso-tank setups in which Langendorff-perfused hearts (n = 4) were suspended and potentials recorded simultaneously from torso and epicardial surfaces. 49 different signal processing methods were applied to torso potentials, grouped as i) high-frequency noise removal (HFR) methods ii) baseline drift removal (BDR) methods and iii) combined HFR+BDR. The inverse problem was solved and reconstructed electrograms and activation maps compared to those directly recorded.Results:HFR showed no difference compared to not filtering in terms of absolute differences in reconstructed electrogram amplitudes nor median correlation in QRS waveforms (p > 0.05). However, correlation and mean absolute error of activation times and pacing site localization were improved with all methods except a notch filter. HFR applied post-reconstruction produced no differences compared to pre-reconstruction. BDR and BDR+HFR significantly improved absolute and relative difference, and correlation in electrograms (p < 0.05). While BDR+HFR combined improved activation time and pacing site detection, BDR alone produced significantly lower correlation and higher localization errors (p < 0.05).Conclusion:BDR improves reconstructed electrogram morphologies and amplitudes due to a reduction in lambda value selected for the inverse problem. The simplest method (resetting the isoelectric point) is sufficient to see these improvements. HFR does not impact electrogram accuracy, but does impact post-processing to extract features such as activation times. Removal of line noise is insufficient to see these changes. HFR should be applied post-reconstruction to ensure over-filtering does not occur.

Highlights

N ONINVASIVE electrocardiographic imaging (ECGI) has been developed to provide high-resolution images of cardiac electrical activity
While there was a significant difference in lambda values between high-frequency noise removal (HFR) methods, the absolute differences were minimal
This study has demonstrated the impact of different signal processing methods on an epicardial potential based inverse method using four data sets from two distinct experimental setups

Summary

Introduction

N ONINVASIVE electrocardiographic imaging (ECGI) has been developed to provide high-resolution images of cardiac electrical activity. ECGI is increasingly being used to guide ablation therapy, such as in the identification of the origin of premature ventricular contractions (PVCs) or epicardial exit sites of ventricular arrhythmias [1]–[4]. Despite the increase in clinical adoption, previous validation studies of epicardial potential-based methods have shown varying results with respect to accuracy. Recent clinical validation studies have shown large variability in activation map reconstruction accuracy (correlation from −0.68 to 0.82 in one study [9], 0.29 to 0.80 in another [10]). The variability in accuracy seen between different centers may be linked to the different inverse or post-processing methods used by each group. Given that large variability exists even in single center studies using the same inverse method pipeline, it is unlikely the inverse or post-processing methods are the sole source of this variability

Objectives

Results

Conclusion