Non-invasive prediction of catheter ablation outcome in persistent atrial fibrillation by exploiting the spatial diversity of surface ECG

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.

Early differentiation of long-standing persistent atrial fibrillation using the characteristics of fibrillatory waves in surface ECG multi-leads

We characterized the f-waves in atrial fibrillation (AF) in the surface ECG by quantifying the amplitude, irregularity, and dominant rate of the f-waves in leads II, aVL, and V₁, and investigated whether those parameters of the f-waves could discriminate long-standing persistent AF (LPeAF) from non-LPeAF. A total of 224 AF patients were enrolled: 112 with PAF (87 males), 48 with PeAF (38 males), and 64 with LPeAF (47 males). The f-waves in surface ECG leads V₁, aVL, and II, which reflect well electrical activity in the right atrium (RA), the left atrium (LA), and both atria, respectively, were analyzed. The f-waves for LPeAF had lower amplitudes in II and aVL, increased irregularity and a higher dominant rate in II and V₁ compared to PAF and PeAF (all p < 0.02). In a multivariate analysis, a low amplitude in lead II (<34.6 uV) and high dominant rate in lead V₁ (≧390/min) (p < 0.00₁) independently discriminated LPeAF from the other AF types. The f-waves combined with both a low amplitude in lead II and high dominant rate in lead V₁ were significantly associated with LPeAF (OR 6.27, p < 0.001). Characteristics of the f-waves on the surface ECG could discriminate LPeAF from other types of AF.

Evaluating the Atrial Myopathy Underlying Atrial Fibrillation: Identifying the Arrhythmogenic and Thrombogenic Substrate

Atrial disease or myopathy forms the substrate for atrial fibrillation (AF) and underlies the potential for atrial thrombus formation and subsequent stroke. Current diagnostic approaches in patients with AF focus on identifying clinical predictors with the evaluation of left atrial size by echocardiography serving as the sole measure specifically evaluating the atrium. Although the atrial substrate underlying AF is likely developing for years before the onset of AF, there is no current evaluation to identify the preclinical atrial myopathy. Atrial fibrosis is 1 component of the atrial substrate that has garnered recent attention based on newer MRI techniques that have been applied to visualize atrial fibrosis in humans with prognostic implications regarding the success of treatment. Advanced ECG signal processing, echocardiographic techniques, and MRI imaging of fibrosis and flow provide up-to-date approaches to evaluate the atrial myopathy underlying AF. Although thromboembolic risk is currently defined by clinical scores, their predictive value is mediocre. Evaluation of stasis via imaging and biomarkers associated with thrombogenesis may provide enhanced approaches to assess risk for stroke in patients with AF. Better delineation of the atrial myopathy that serves as the substrate for AF and thromboembolic complications might improve treatment outcomes. Furthermore, better delineation of the pathophysiologic mechanisms underlying the development of the atrial substrate for AF, particularly in its earlier stages, could help identify blood and imaging biomarkers that could be useful to assess risk for developing new-onset AF and suggest specific pathways that could be targeted for prevention.

Multidimensional characterization of fibrillatory wave amplitude on surface ECG to describe catheter ablation impact on persistent atrial fibrillation

Radiofrequency catheter ablation (CA) is increasingly employed to treat persistent atrial fibrillation (AF). Nevertheless, its success is not always guaranteed, as selection of patients who could positively respond to this therapy does not rely on systematic criteria and still remains an open issue. Moreover, very little is known about the quantitative effects of this treatment over AF electrophysiology, so their quantitative evaluation is not a trivial task. In this contribution, ablation impact is quantified by a descriptor of fibrillatory wave (f-wave) amplitude, so far regarded as a predictor of short-term CA outcome. By means of principal component analysis (PCA), surface electrocardiogram (ECG) spatial diversity is exploited and contributions from all leads are combined to describe average f-wave peak-to-peak amplitude, whose value is automatically computed by an algorithm based on cubic spline interpolation. Our work demonstrates how CA influences f-wave amplitude during the procedure as quantified by ECG inter-lead spatial variability. In addition, we show how such variations depend on procedural outcome and the duration of the postoperative blanking period.