Comparative study of methods for ventricular activity cancellation in atrial electrograms of atrial fibrillation

Digital biomarkers and algorithms for detection of atrial fibrillation using surface electrocardiograms: A systematic review

AIMS

Automated detection of atrial fibrillation (AF) in continuous rhythm registrations is essential in order to prevent complications and optimize treatment of AF. Many algorithms have been developed to detect AF in surface electrocardiograms (ECGs) during the past few years. The aim of this systematic review is to gain more insight into these available classification methods by discussing previously used digital biomarkers and algorithms and make recommendations for future research.

METHODS

On the 14^th of September 2020, the PubMed database was searched for articles focusing on algorithms for AF detection in ECGs using the MeSH terms Atrial Fibrillation, Electrocardiography and Algorithms. Articles which solely focused on differentiation of types of rhythm disorders or prediction of AF termination were excluded.

RESULTS

The search resulted in 451 articles, of which 130 remained after full-text screening. Not only did the amount of research on methods for AF detection increase over the past years, but a trend towards more complex classification methods is observed. Furthermore, three different types of features can be distinguished: atrial features, ventricular features, and signal features. Although AF is an atrial disease, only 22% of the described methods use atrial features.

CONCLUSION

More and more studies focus on improving accuracy of classification methods for AF in ECGs. As a result, algorithms become increasingly complex and less well interpretable. Only a few studies focus on detecting atrial activity in the ECG. Developing innovative methods focusing on detection of atrial activity might provide accurate classifiers without compromising on transparency.

Unipolar electrogram-based voltage mapping with far-field cancellation to improve detection of abnormal atrial substrate during atrial fibrillation

INTRODUCTION

An important substrate for atrial fibrillation (AF) is fibrotic atrial myopathy. Identifying low voltage, myopathic regions during AF using traditional bipolar voltage mapping is limited by the directional dependency of wave propagation. Our objective was to evaluate directionally independent unipolar voltage mapping, but with far-field cancellation, to identify low-voltage regions during AF.

METHODS

In 12 patients undergoing pulmonary vein isolation for AF, high-resolution voltage mapping was performed in the left atrium during sinus rhythm and AF using a roving 20-pole circular catheter. Bipolar electrograms (EGMs) (Bi) < 0.5 mV in sinus rhythm identified low-voltage regions. During AF, bipolar voltage and unipolar voltage maps were created, the latter with (uni-res) and without (uni-orig) far-field cancellation using a novel, validated least-squares algorithm.

RESULTS

Uni-res voltage was ~25% lower than uni-orig for both low voltage and normal atrial regions. Far-field EGM had a dominant frequency (DF) of 4.5-6.0 Hz, and its removal resulted in a lower DF for uni-orig compared with uni-res (5.1 ± 1.5 vs. 4.8 ± 1.5 Hz; p < .001). Compared with Bi, uni-res had a significantly greater area under the receiver operator curve (0.80 vs. 0.77; p < .05), specificity (86% vs. 76%; p < .001), and positive predictive value (43% vs. 30%; p < .001) for detecting low-voltage during AF. Similar improvements in specificity and positive predictive value were evident for uni-res versus uni-orig.

CONCLUSION

Far-field EGM can be reliably removed from uni-orig using our novel, least-squares algorithm. Compared with Bi and uni-orig, uni-res is more accurate in detecting low-voltage regions during AF. This approach may improve substrate mapping and ablation during AF, and merits further study.

Refined Ventricular Activity Cancellation in Electrograms During Atrial Fibrillation by Combining Average Beat Subtraction and Interpolation

Many techniques have been developed to cancel the ventricular interference in atrial electrograms (AEG) during atrial fibrillation. In particular, average beat subtraction (ABS) and interpolation are among those mostly adopted. However, ABS usually leaves high power residues and discontinuity at the borders, whereas interpolation totally substitutes the residual activity with a forecasting that might fail at the center of the cancellation segment. In this study, we proposed a new algorithm to refine the ventricular estimate provided by ABS, in such a way that the residual activity should likely be distributed as the local atrial activity. Briefly, the local atrial activity is first modeled with an autoregressive (AR) process, then the estimate is refined by maximizing the log likelihood of the atrial residual activity according to the fitted AR model. We tested the new algorithm on both synthetic and real AEGs, and compared the performance with other four algorithms (two variants of ABS, interpolation and zero substitution). On synthetic data, our algorithm outperformed all the others in terms of average root mean square error (0.043 vs 0.046 for interpolation; p <; 0.05). On real data, our methodology outperformed two variants of ABS (p <; 0.05) and performed similarly to interpolation when considering the high power residues left (both <; 5%), and the log likelihood with the fitted AR model.

Reference database and performance evaluation of methods for extraction of atrial fibrillatory waves in the ECG

OBJECTIVE

This study proposes a reference database, composed of a large number of simulated ECG signals in atrial fibrillation (AF), for investigating the performance of methods for extraction of atrial fibrillatory waves (f-waves).

APPROACH

The simulated signals are produced using a recently published and validated model of 12-lead ECGs in AF. The database is composed of eight signal sets together accounting for a wide range of characteristics known to represent major challenges in f-wave extraction, including high heart rates, high morphological QRST variability, and the presence of ventricular premature beats. Each set contains 30 5 min signals with different f-wave amplitudes. The database is used for the purpose of investigating the statistical association between different indices, designed for use with either real or simulated signals.

MAIN RESULTS

Using the database, available at the PhysioNet repository of physiological signals, the performance indices unnormalized ventricular residue (uVR), designed for real signals, and the root mean square error, designed for simulated signals, were found to exhibit the strongest association, leading to the recommendation that uVR should be used when characterizing performance in real signals.

SIGNIFICANCE

The proposed database facilitates comparison of the performance of different f-wave extraction methods and makes it possible to express performance in terms of the error between simulated and extracted f-wave signals.

Dynamics of AV coupling during human atrial fibrillation: role of atrial rate

The causal relationship between atrial and ventricular activities during human atrial fibrillation (AF) is poorly understood. This study analyzed the effects of an increase in atrial rate on the link between atrial and ventricular activities during AF. Atrial and ventricular time series were determined in 14 patients during the spontaneous acceleration of the atrial rhythm at AF onset. The dynamic relationship between atrial and ventricular activities was quantified in terms of atrioventricular (AV) coupling by AV synchrogram analysis. The technique identified n: m coupling patterns ( n atrial beats in m ventricular cycles), quantifying their percentage, maximal length, and conduction ratio (= m/ n). Simulations with a difference-equation AV model were performed to correlate the observed dynamics to specific atrial/nodal properties. The atrial rate increase significantly affected AV coupling and ventricular response during AF. The shortening of atrial intervals from 185 ± 32 to 165 ± 24 ms ( P < 0.001) determined transitions toward AV patterns with progressively decreasing m/ n ratios (from conduction ratio = 0.34 ± 0.09 to 0.29 ± 0.08, P < 0.01), lower occurrence (from percentage of coupled beats = 27.1 ± 8.0 to 21.8 ± 6.9%, P < 0.05), and higher instability (from maximal length = 3.9 ± 1.5 to 2.8 ± 0.7 s, P < 0.01). Advanced levels of AV block and coupling instability at higher atrial rates were associated with increased ventricular interval variability (from 123 ± 52 to 133 ± 55 ms, P < 0.05). AV pattern transitions and coupling instability in patients were predicted, assuming the filtering of high-rate irregular atrial beats by the slow recovery of nodal excitability. These results support the role of atrial rate in determining AV coupling and ventricular response and may have implications for rate control in AF.

The logical operator map identifies novel candidate markers for critical sites in patients with atrial fibrillation

The identification of suitable markers for critical patterns during atrial fibrillation (AF) may be crucial to guide an effective ablation treatment. Single parameter maps, based on dominant frequency and complex fractionated electrograms, have been proposed as a tool for electrogram-guided ablation, however the specificity of these markers is debated. Experimental studies suggest that AF critical patterns may be identified on the basis of specific rate and organization features, where rapid organized and rapid fragmented activities characterize respectively localized sources and critical substrates. In this paper we introduce the logical operator map, a novel mapping tool for a point-by-point identification and localization of AF critical sites. Based on advanced signal and image processing techniques, the approach combines in a single map electrogram-derived rate and organization features with tomographic anatomical detail. The construction of the anatomically-detailed logical operator map is based on the time-domain estimation of atrial rate and organization in terms of cycle length and wave-similarity, the logical combination of these indexes to obtain suitable markers of critical sites, and the multimodal integration of electrophysiological and anatomical information by segmentation and registration techniques. Logical operator maps were constructed in 14 patients with persistent AF, showing the capability of the combined rate and organization markers to identify with high selectivity the subset of electrograms associated with localized sources and critical substrates. The precise anatomical localization of these critical sites revealed the confinement of rapid organized sources in the left atrium with organization and rate gradients towards the surrounding tissue, and the presence of rapid fragmented electrograms in proximity of the sources. By merging in a single map the most relevant electrophysiological and anatomical features of the AF process, the logical operator map may have significant clinical impact as a direct, comprehensive tool to understand arrhythmia mechanisms in the single patient and guide more conservative, step-wise ablation.

Ventricular activity cancellation in electrograms during atrial fibrillation with constraints on residuals' power

During atrial fibrillation (AF), cancellation of ventricular activity from atrial electrograms (AEG) is commonly performed by template matching and subtraction (TMS): a running template, built in correspondence of QRSs, is subtracted from the AEG to uncover atrial activity (AA). However, TMS can produce poor cancellation, leaving high-power residues. In this study, we propose to modulate the templates before subtraction, in order to make the residuals as similar as possible to the nearby atrial activity, avoiding high-power ones. The coefficients used to modulate the template are estimated by maximizing, via Multi-swarm Particle Swarm Optimization, a fitness function. The modulated TMS method (mTMS) was tested on synthetic and real AEGs. Cancellation performances were assessed using: normalized mean squared error (NMSE, computed on simulated data only), reduction of ventricular activity (VDR), and percentage of segments (PP) whose power was outside the standard range of the atrial power. All testings suggested that mTMS is an improvement over TMS alone, being, on simulated data, NMSE and PP significantly decreased while VDR significantly increased. Similar results were obtained on real electrograms (median values of CS1 recordings PP: 2.44 vs. 0.38 p < 0.001; VDR: 6.71 vs. 8.15 p < 0.001).

Dynamic time warping applied to estimate atrial fibrillation temporal organization from the surface electrocardiogram

Atrial fibrillation (AF) is the most commonly diagnosed arrhythmia in clinical practice. However, the mechanisms responsible for its induction and maintenance still are not fully understood. To this respect, analysis of the electrical activity organization within the atria could play an important role in their proper interpretation. Although many algorithms to quantify AF organization from invasive electrograms can be found in the literature, a reduced number of indirect estimators from the standard ECG have been proposed to date. Furthermore, these surface methods can only yield a global AF organization assessment, blurring the possible information that each individual fibrillatory (f) wave may provide. To this respect, the present manuscript proposes a novel method for direct and short-time AF organization estimation from single-lead surface ECG recordings. Through the computation of morphological variations among f waves, the temporal arrhythmia organization is estimated. The f waves are individually extracted and delineated from the atrial activity signal, making use of a dynamic time warping approach. The proposed algorithm was tested on real AF surface recordings in order to discriminate atrial signals with different organization degrees, obtaining a diagnostic accuracy higher than 88%. In addition, its performance was validated by comparison with two temporal organization measures from invasive unipolar electrograms of both atria, providing statistically significant linear correlations between invasive and non-invasive estimates. As a consequence, new standpoints are opened through this work in the non-invasive analysis of AF, where the individualized study of each f wave could assess short-time AF organization, would improve the understanding of AF mechanisms and become useful for its clinical treatment.

Novel methods for characterization of paroxysmal atrial fibrillation in human left atria

INTRODUCTION

More effective methods for characterizing 3D electrical activity in human left atrium (LA) are needed to identify substrates/triggers and microreentrant circuit for paroxysmal atrial fibrillation (PAF). We describe a novel wavelet-based approach and wave-front centroid tracking that have been used to reconstruct regional activation frequency and electrical activation pathways from non-contact multi-electrode array.

METHODS

Data from 13 patients acquired prior to ablation for PAF with a 64 electrode noncontact catheter positioned in the LA were analysed. Unipolar electrograms were reconstructed at 2048 locations across each LA endocardial surface. Weighted fine- and coarse-scale electrograms were constructed by wavelet decomposition and combined with peak detection to identify atrial fibrillation (AF) activation frequency and fractionated activity at each site. LA regions with upper quartile AF frequencies were identified for each patient. On the other hand, a wave-front centroid tracking approach was introduced for this first time to detect macro-reentrant circuit during PAF.

RESULTS

The results employing wavelet-based analysis on atrial unipolar electrograms are validated by the signals recorded simultaneously via the contacted ablation catheter and visually tracking the 3D spread of activation through the interest region. Multiple connected regions of high frequency electrical activity were seen; most often in left superior pulmonary vein (10/12), septum (9/12) and atrial roof (9/12), as well as the ridge (8/12). The wave-front centroid tracking approach detects a major macro circuit involving LPVs, PLA, atrial floor, MV, septum, atrial roof and ridge. The regions with high frequency by wave-front tracking are consistent with the results using wavelet approach and our clinical observations.

CONCLUSIONS

The wavelet-based technique and wave-front centroid tracking approach provide a robust means of extracting spatio-temporal characteristics of PAF. The approach could facilitate accurate identification of pro-arrhythmic substrate and triggers, and therefore, to improve success rate of catheter ablation for AF.

An efficient method for ectopic beats cancellation based on radial basis function

The analysis of the surface Electrocardiogram (ECG) is the most extended noninvasive technique in cardiological diagnosis. In order to properly use the ECG, we need to cancel out ectopic beats. These beats may occur in both normal subjects and patients with heart disease, and their presence represents an important source of error which must be handled before any other analysis. This paper presents a method for electrocardiogram ectopic beat cancellation based on Radial Basis Function Neural Network (RBFNN). A train-able neural network ensemble approach to develop customized electrocardiogram beat classifier in an effort to further improve the performance of ECG processing and to offer individualized health care is presented. Six types of beats including: Normal Beats (NB); Premature Ventricular Contractions (PVC); Left Bundle Branch Blocks (LBBB); Right Bundle Branch Blocks (RBBB); Paced Beats (PB) and Ectopic Beats (EB) are obtained from the MIT-BIH arrhythmia database. Four morphological features are extracted from each beat after the preprocessing of the selected records. Average Results for the RBFNN based method provided an ectopic beat reduction (EBR) of (mean ± std) EBR = 7, 23 ± 2.18 in contrast to traditional compared methods that, for the best case, yielded EBR = 4.05 ± 2.13. The results prove that RBFNN based methods are able to obtain a very accurate reduction of ectopic beats together with low distortion of the QRST complex.

Electropathological substrate detection of persistent atrial fibrillation--a novel method to analyze unipolar electrograms of noncontact mapping

Radiofrequency catheter ablation as a curative method for atrial fibrillation (AF) has become increasingly popular. Patients with paroxysmal AF have been treated by catheter ablation with great success, but so far this treatment has been less effective for patients with persistent AF. Usually there are multiple triggers or substrates during persistent AF and their exact locations are unclear. On the other hand, the non-contact mapping system (Ensite 3000, St Jude Medical) producing thousands of virtual endocardial electrograms, has gradually become accepted as a powerful tool to use on patients before and after ablation. Effective mathematical tools to detect the substrates of AF from unipolar electrograms produced by the non-contact mapping are few, though many methods are available for performing this task with bipolar electrograms. In this work, we introduce for the first time a simple and efficient approach to automatically and systematically determine the substrate of persistent AF in order to guide catheter ablation via the non-contact mapping.

A ventricular activity cancellation algorithm based on event synchronous adaptive filter for single-lead electrocardiograms

Recently, it has become very important to analyze atrial activity (AA) and to detect arrhythmic AAs and, for this, complete ventricular activity (VA) cancellation is prerequisite. There have been several VA cancellation algorithms for multi-lead ECG but VA cancellation algorithm for single-lead is quite a few. In this study, we have modeled thoracic ECG and, based on this model, proposed a novel VA cancellation algorithm based on event synchronous adaptive filter (ESAF). In this ESAF, the AF ECG was treated as a primary input and event-synchronous impulse train (ESIT) as a reference. And, ESIT was generated so to be synchronized with the ventricular activity by detecting QRS complex. To evaluate the performance, it was applied to the AA estimation problem in atrial fibrillation electrocardiograms. As results, even with low computational cost, this ESAF based algorithm showed better performance than the ABS method and comparable performance to algorithm based on PCA (principal component analysis) or SVD (singular value decomposition). We also proposed an expanded version of ESAF for some AF ECGs with bimorphic VAs and this also showed reasonable performance. Ultimately, our proposed algorithm was found to estimate AA precisely even though it is possible to implement in real-time. We expect our algorithm to replace the most widely used method, that is, the ABS (averaged beat subtraction) method.

Validation of surface atrial fibrillation organization indicators through invasive recordings

Studies related to atrial fibrillation (AF) have shown that surface lead V(1) reflects mainly the dominant atrial frequency (DAF) of the right atrium (RA), which has been widely used to analyze this arrhythmia. AF organization and fibrillatory (f) waves amplitude are two recently proposed non-invasive AF markers, which have not been yet validated with invasive recordings. In this work, these two non-invasive metrics have been compared with similar measures recorded from two unipolar atrial electrograms (AEGs). Results obtained from 38 patients showed statistically significant correlations between the values measured from surface and invasive recordings, thus corroborating the usefulness of the aforesaid markers in the non-invasive study of AF. Precisely, for AF organization, the correlation coefficients between surface and both AEGs were R = 0.926 (p < 0.001) and R = 0.932 (p < 0.001), respectively. For f waves amplitude, slightly lower significant relationships were noticed, the correlation coefficients being R = 0.765 (p < 0.001) and R = 0.842 (p < 0.001), respectively. These outcomes together with interesting linear relationships found among the parameters, suggest that AF organization and f waves amplitude can characterize non-invasively the epicardial activity related to AF.

A time-domain approach for the identification of atrial fibrillation drivers

The localization of atrial fibrillation (AF) driver sources, characterized by rapid and regular electrical activity, is crucial for an effective ablation treatment. This work proposes a double-criteria approach for the identification of AF drivers based on a time-domain evaluation of atrial rate and AF organization. These two features are quantified by the measurement of atrial cycle length (ACL) and wave-similarity (WS). Based on ACL/WS formalism, AF drivers can be operatively defined as sites displaying electrical activity with high-rate and high-similarity (HR AND HS). The capability of ACL/WS analysis to identify AF driver sites and distinguish them from non-critical areas is shown in representative examples. The double-criteria evaluation for the identification of AF drivers, provided by our time-domain approach, might open new perspectives for the development of electrogram-guided ablation strategies in the single patient.

Adaptive singular value cancelation of ventricular activity in single-lead atrial fibrillation electrocardiograms

The proper analysis and characterization of atrial fibrillation (AF) from surface electrocardiographic (ECG) recordings requires to cancel out the ventricular activity (VA), which is composed of the QRS complex and the T wave. Historically, for single-lead ECGs, the averaged beat subtraction (ABS) has been the most widely used technique. However, this method is very sensitive to QRST wave variations and, moreover, high-quality cancelation templates may be difficult to obtain when only short length and single-lead recordings are available. In order to overcome these limitations, a new QRST cancelation method based on adaptive singular value cancelation (ASVC) applied to each single beat is proposed. In addition, an exhaustive study about the optimal set of complexes for better cancelation of every beat is also presented for the first time. The whole study has been carried out with both simulated and real AF signals. For simulated AF, the cancelation performance was evaluated making use of a cross-correlation index and the normalized mean square error (nmse) between the estimated and the original atrial activity (AA). For real AF signals, two additional new parameters were proposed. First, the ventricular residue (VR) index estimated the presence of ventricular activity in the extracted AA. Second, the similarity (S) evaluated how the algorithm preserved the AA segments out of the QRST interval. Results indicated that for simulated AF signals, mean correlation, nmse, VR and S values were 0.945 +/- 0.024, 0.332 +/- 0.073, 1.552 +/- 0.386 and 0.986 +/- 0.012, respectively, for the ASVC method and 0.866 +/- 0.042, 0.424 +/- 0.120, 2.161 +/- 0.564 and 0.922 +/- 0.051 for ABS. In the case of real signals, the mean VR and S values were 1.725 +/- 0.826 and 0.983 +/- 0.038, respectively, for ASVC and 3.159 +/- 1.097 and 0.951 +/- 0.049 for ABS. Thus, ASVC provides a more accurate beat-to-beat ventricular QRST representation than traditional techniques. As a consequence, VA cancelation is optimized and the AA can be extracted more precisely. Finally, the study has proven that optimal VA cancelation is achieved when a number between 20 and 30 complexes is selected following a correlation-based strategy.

P wave detector with PP rhythm tracking: evaluation in different arrhythmia contexts

Automatic detection of atrial activity (P waves) in an electrocardiogram (ECG) is a crucial task to diagnose the presence of arrhythmias. The P wave is difficult to detect and most of the approaches in the literature have been evaluated on normal sinus rhythms and rarely considered arrhythmia contexts other than atrial flutter and fibrillation. A novel knowledge-based P wave detector algorithm is presented. It is self-adaptive to the patient and able to deal with certain arrhythmias by tracking the PP rhythm. The detector has been tested on 12 records of the MIT-BIH arrhythmia database containing several ventricular and supra-ventricular arrhythmias. On the overall records, the detector demonstrates Se = 96.60% and Pr = 95.46%; for the normal sinus rhythm, it reaches Se = 97.76% and Pr = 96.80% and, in the case of Mobitz type II, it demonstrates Se = 72.79% and Pr = 99.51%. It also shows good performance for trigeminy and bigeminy, and outperforms some more sophisticated techniques. Although the results emphasize the difficulty of P wave detection in difficult arrhythmias (supra and ventricular tachycardias), it shows that domain knowledge can efficiently support signal processing techniques.