The application of nonlinear metrics to assess organization differences in short recordings of paroxysmal and persistent atrial fibrillation

Machine learning based potentiating impacts of 12-lead ECG for classifying paroxysmal versus non-paroxysmal atrial fibrillation

Background

Conventional modality requires several days observation by Holter monitor to differentiate atrial fibrillation (AF) between Paroxysmal atrial fibrillation (PAF) and Non-paroxysmal atrial fibrillation (Non-PAF). Rapid and practical differentiating approach is needed.

Objective

To develop a machine learning model that observes 10-s of standard 12-lead electrocardiograph (ECG) for real-time classification of AF between PAF versus Non-PAF.

Methods

In this multicenter, retrospective cohort study, the model training and cross-validation was performed on a dataset consisting of 741 patients enrolled from Severance Hospital, South Korea. For cross-institutional validation, the trained model was applied to an independent data set of 600 patients enrolled from Ewha University Hospital, South Korea. Lasso regression was applied to develop the model.

Results

In the primary analysis, the Area Under the Receiver Operating Characteristic Curve (AUC) on the test set for the model that predicted AF subtype only using ECG was 0.72 (95% CI 0.65–0.80). In the secondary analysis, AUC only using baseline characteristics was 0.53 (95% CI 0.45–0.61), while the model that employed both baseline characteristics and ECG parameters was 0.72 (95% CI 0.65–0.80). Moreover, the model that incorporated baseline characteristics, ECG, and Echocardiographic parameters achieved an AUC of 0.76 (95% CI 0.678–0.855) on the test set.

Conclusions

Our machine learning model using ECG has potential for automatic differentiation of AF between PAF versus Non-PAF achieving high accuracy. The inclusion of  Echocardiographic parameters further increases model performance. Further studies are needed to clarify the next steps towards clinical translation of the proposed algorithm.

Classification of De novo post-operative and persistent atrial fibrillation using multi-channel ECG recordings

Atrial fibrillation (AF) is the most sustained arrhythmia in the heart and also the most common complication developed after cardiac surgery. Due to its progressive nature, timely detection of AF is important. Currently, physicians use a surface electrocardiogram (ECG) for AF diagnosis. However, when the patient develops AF, its various development stages are not distinguishable for cardiologists based on visual inspection of the surface ECG signals. Therefore, severity detection of AF could start from differentiating between short-lasting AF and long-lasting AF. Here, de novo post-operative AF (POAF) is a good model for short-lasting AF while long-lasting AF can be represented by persistent AF. Therefore, we address in this paper a binary severity detection of AF for two specific types of AF. We focus on the differentiation of these two types as de novo POAF is the first time that a patient develops AF. Hence, comparing its development to a more severe stage of AF (e.g., persistent AF) could be beneficial in unveiling the electrical changes in the atrium. To the best of our knowledge, this is the first paper that aims to differentiate these different AF stages. We propose a method that consists of three sets of discriminative features based on fundamentally different aspects of the multi-channel ECG data, namely based on the analysis of RR intervals, a greyscale image representation of the vectorcardiogram, and the frequency domain representation of the ECG. Due to the nature of AF, these features are able to capture both morphological and rhythmic changes in the ECGs. Our classification system consists of a random forest classifier, after a feature selection stage using the ReliefF method. The detection efficiency is tested on 151 patients using 5-fold cross-validation. We achieved 89.07% accuracy in the classification of de novo POAF and persistent AF. The results show that the features are discriminative to reveal the severity of AF. Moreover, inspection of the most important features sheds light on the different characteristics of de novo post-operative and persistent AF.

Gabor frames for classification of paroxysmal and persistent atrial fibrillation episodes

In this study, we propose a new classification method for early differentiation of paroxysmal and persistent atrial fibrillation episodes, i.e. those which spontaneously or with external intervention will return to sinus rhythm within 7 days of onset from the ones where the arrhythmia is sustained for more than 7 days. Today, clinicians provide patients classification once the course of the arrhythmia has been disclosed. This classification problem is dealt with in this study. We study a sparse representation of surface electrocardiogram signals by means of Gabor frames and afterwards we apply a linear discriminant analysis. Thus, we provide an early discrimination, obtaining promising performances on a heterogeneous cohort of patients in terms of pharmacological treatment and state of progression of the arrhythmia: 95% sensitivity, 82% specificity, 89% accuracy. In this manner, the proposed method can help clinicians to choose the most appropriate treatment using the electrocardiogram, which is a widely available and non-invasive technique. This early differentiation is clinically highly significant in order to choose optimal patients who may undergo catheter ablation with higher success rates.

Intracardiac impedance response during acute AF internal cardioversion using novel rectilinear and capacitor-discharge waveforms

Intracardiac impedance (ICI) is a major determinant of success during internal cardioversion of atrial fibrillation (AF). However, there have been few studies that have examined the dynamic behaviour of atrial impedance during internal cardioversion in relation to clinical outcome. In this study, voltage and current waveforms captured during internal cardioversion of acute AF in ovine models using novel radiofrequency (RF) generated low-tilt rectilinear and conventional capacitor-discharge based shock waveforms were retrospectively analysed using a digital signal processing algorithm to investigate the dynamic behaviour of atrial impedance during cardioversion. The algorithm was specifically designed to facilitate the simultaneous analysis of multiple impedance parameters, including: mean intracardiac impedance (Z M), intracardiac impedance variance (ICIV) and impedance amplitude spectrum area (IAMSA) for each cardioversion event. A significant reduction in ICI was observed when comparing two successive shocks of increasing energy where cardioversion outcome was successful. In addition, ICIV and IAMSA variables were found to inversely correlate to the magnitude of energy delivered; with a stronger correlation found to the former parameter. In conclusion, ICIV and IAMSA have been evidenced as two key dynamic intracardiac impedance variables that may prove useful in better understanding of the cardioversion process and that could potentially act as prognostic markers with respect to clinical outcome.

Electrocardiographic Spectral Features for Long-Term Outcome Prognosis of Atrial Fibrillation Catheter Ablation

Atrial fibrillation (AF) is the most common arrhythmia in routine clinical practice. Despite many years of research, its mechanisms still are not well understood, thus reducing the effectiveness of AF treatments. Nowadays, pulmonary vein isolation by catheter ablation is the treatment of choice for AF resistant either to pharmacological or electrical cardioversion. However, given that long-term recurrences are common, an appropriate patient selection before the procedure is of paramount relevance in the improvement of AF catheter ablation outcome. The present work studies how several spectral features of the atrial activity (AA) from a single lead of the surface electrocardiogram (ECG) can become potential pre-ablation predictors of long-term (>2 months) sinus rhythm maintenance. Among all the analyzed spectral features, results indicated that the most significant single predictor of paroxysmal AF ablation treatment outcome was related to the amplitude of the first harmonic of the dominant frequency, providing sensitivity (Se), specificity (Sp) and accuracy (Ac) values of 90%, 42.86 and 77.78%, respectively. On the other hand, the AA harmonic structure was the most significant single predictor for persistent AF, with Se, Sp and Ac values of 100%, 54.55 and 77.27%, respectively. A logistic regression analysis, mainly based on spectral amplitudes as well as on the harmonic structure of the AA, provided a higher predictive ability both for paroxysmal AF (Se = 100%, Sp = 57.14% and Ac = 88.89%) and persistent AF (Se = 90.91%, Sp = 72.73 and Ac = 81.82%). In conclusion, the study of key AA spectral features from the surface ECG can provide a significant preoperative prognosis of AF catheter ablation outcome at long-term follow-up.

Nonlinear analysis of right atrial electrograms predicts termination of persistent atrial fibrillation within the left atrium by catheter ablation

The termination of long-standing persistent atrial fibrillation (LS-pAF) can be achieved by stepwise catheter ablation (step-CA) within the left atrium (LA). Our study aims to applying complexity measures derived from nonlinear time series analysis in order to characterize LS-pAF in terms of organization and to identify patients in whom AF can be terminated from those in whom AF cannot be terminated by step-CA within the LA. A total of 33 consecutive patients (age 61  ±  7 years, sustained AF duration 19  ±  11 months) with LS-pAF underwent step-CA. The organization of right bipolar electrograms before and during the ablation procedure was assessed using the coarse-grained correlation dimension. LS-pAF was terminated into sinus rhythm or atrial tachycardia in 22 patients during step-CA within the LA (left terminated patients-LT). In 11 patients the ablation procedure failed to terminate AF within LA (not left terminated patients-NLT). The statistical analysis of the estimated coarse-grained correlation dimension revealed that a higher right atrial (RA) organization before step-CA was associated to AF termination within the LA. During the ablation procedure, the level of RA organization displayed distinctive evolution between LT and NLT patients with a significant organization increase before AF termination for the LT patients.

Measures of spatiotemporal organization differentiate persistent from long-standing atrial fibrillation

AIMS

This study presents an automatic diagnostic method for the discrimination between persistent and long-standing atrial fibrillation (AF) based on the surface electrocardiogram (ECG).

METHODS AND RESULTS

Standard 12-lead ECG recordings were acquired in 53 patients with either persistent (N= 20) or long-standing AF (N= 33), the latter including both long-standing persistent and permanent AF. A combined frequency analysis of multiple ECG leads followed by the computation of standard complexity measures provided a method for the quantification of spatiotemporal AF organization. All possible pairs of precordial ECG leads were analysed by this method and resulting organization measures were used for automatic classification of persistent and long-standing AF signals. Correct classification rates of 84.9% were obtained, with a predictive value for long-standing AF of 93.1%. Spatiotemporal organization as measured in lateral precordial leads V5 and V6 was shown to be significantly lower during long-standing AF than persistent AF, suggesting that time-related alterations in left atrial electrical activity can be detected in the ECG.

CONCLUSION

Accurate discrimination between persistent and long-standing AF based on standard surface recordings was demonstrated. This information could contribute to optimize the management of sustained AF, permitting appropriate therapeutic decisions and thereby providing substantial clinical cost savings.

Application of frequency and sample entropy to discriminate long-term recordings of paroxysmal and persistent atrial fibrillation

Atrial fibrillation (AF) is the most common arrhythmia in clinical practice. At an early stage of the disease, AF may terminate spontaneously and is then referred to as paroxysmal AF. On the other hand, when external intervention is required for the arrhythmia to terminate, it is referred to as persistent AF. In this work, a method to discriminate between paroxysmal and persistent AF in the long-term ECGs is presented. The dominant frequency as well as the organization of the atrial activity are employed to characterize AF. The dominant atrial frequency (DAF) is estimated using hidden Markov model based frequency tracking, and organization is estimated by the sample entropy of the main atrial wave (MAW) and the first two harmonics, respectively. Long-term variations in DAF and organization from 50 ECG recordings were evaluated, showing that episodes of paroxysmal AF were consistently associated with lower DAF and organization of the MAW and the harmonics, than was persistent AF. Discrimination of paroxysmal and persistent AF resulted in classification rates of 84.1±26.1%, thus suggesting that it possible to discriminate between paroxysmal and persistent AF in patients without previously known AF history.

Surface ECG organization time course analysis along onward episodes of paroxysmal atrial fibrillation

The complete understanding of the mechanisms leading to the initiation, maintenance and self-termination of atrial fibrillation (AF) still is an unsolved challenge for cardiac electrophysiology. Studies in which AF has been induced have shown that electrophysiological and structural remodeling of the atria during the arrhythmia could play an important role in the transition from paroxysmal to persistent AF. However, to this day, the time course of the atrial remodeling along onward episodes of non-induced paroxysmal AF has not been investigated yet. In this work, a non-invasive method, based on the regularity estimation of AF through sample entropy (SampEn), has been used to assess the organization evolution along onward episodes of paroxysmal AF. Given that AF organization has been associated to the number of existing wavelets wandering throughout the atrial tissue, SampEn could be considered as a concomitant estimator of atrial remodeling. The achieved results, in close agreement with previous findings obtained from invasive recordings, proved several relevant aspects of arial remodeling. Firstly, a progressive disorganization increase (SampEn increase) along onward episodes of AF has been observed for 63% of the analyzed patients, whereas a stable AF organization degree has been appreciated in the remaining 37%. Next, a positive correlation between episode duration and SampEn has been obtained (R=0.541, p<0.01). Finally, a remarkable influence of the fibrillation-free interval, preceding each episode, on the corresponding level of AF organization at the onset of the subsequent AF episode has been observed, with a correlation between these two indices of R=0.389 (p<0.01). As a consequence, it could be considered that atrial electrophysiological dynamics that occur along onward paroxysmal AF episodes are reflected and can be quantified from ECG recordings through non-invasive organization estimation.

Noninvasive organization analysis along consecutive episodes of paroxysmal atrial fibrillation

Atrial fibrillation (AF) is the most common cardiac arrhythmia in clinical practice. Although its mechanisms are incompletely understood, electrophysiological and structural remodeling of the atria seem to play an important role in the arrhythmia transition from paroxysmal to persistent. However, the time course of the atrial remodeling along onward episodes of non-induced paroxysmal AF has not been investigated yet. In this work, a non-invasive method, based on the regularity estimation of AF through sample entropy (SampEn), has been used to assess the organization evolution along onward episodes of paroxysmal AF. Given that AF organization has been associated to the number of existing wavelets wandering throughout the atrial tissue, SampEn could be considered as a concomitant estimator of atrial remodeling. The achieved results, in close agreement with previous findings obtained from invasive recordings, showed a progressive disorganization increase along onward episodes of AF for 63% of the analyzed patients and a stable AF organization degree in the remaining 37%. Additionally, a positive correlation between episode duration and SampEn was also noticed (R = 0.541, p < 0.01). As a consequence, it could be considered that atrial electrophysiological dynamics that occur along onward paroxysmal AF episodes are reflected and can be quantified from ECG recordings through non-invasive organization estimation.