Predicting the spontaneous termination of atrial fibrillation based on Poincare section in the electrocardiogram phase space

Abstract

Atrial fibrillation (AF) is a commonly encountered cardiac arrhythmia. Predicting the conditions under which AF terminates spontaneously is an important task that would bring great benefit to both patients and clinicians. In this study, a new method was proposed to predict spontaneous AF termination by employing the points of section (POS) coordinates along a Poincare section in the electrocardiogram (ECG) phase space. The AF Termination Database provided by PhysioNet for the Computers in Cardiology Challenge 2004 was applied in the present study. It includes one training dataset and two testing datasets, A and B. The present investigation was initiated by producing a two-dimensional reconstructed phase space (RPS) of the ECG. Then, a Poincare line was drawn in a direction that included the maximum point distribution in the RPS and also passed through the origin of the RPS coordinate system. Afterward, the coordinates of the RPS trajectory intersections with this Poincare line were extracted to capture the local behavior related to the arrhythmia under investigation. The POS corresponding to atrial activity were selected with regard to the fact that similar ECG morphologies such as P waves, which are corresponding to atrial activity, distribute in a specific region of the RPS. Thirteen features were extracted from the selected intersection points to quantify their distributions. To select the best feature subset, a genetic algorithm (GA), in combination with a support vector machine (SVM), was applied to the training dataset. Based on the selected features and trained SVM, the performance of the proposed method was evaluated using the testing datasets. The results showed that 86.67% of dataset A and 80% of dataset B were correctly classified. This classification accuracy is in the same range as or higher than that of recent studies in this area. These results show that the proposed method, in which no complicated QRST cancelation algorithm was used, has the potential to predict AF termination.

Analyzing single-lead short ECG recordings using dense convolutional neural networks and feature-based post-processing to detect atrial fibrillation

“…The false-nearest neighbor algorithm (FNN) is a technique that quantifies the points which appear close together in a low dimension that are actually far apart in a RPS with higher dimension (Kennel et al 1992). Using FNN and an area enclosed by the outer loop (Parvaneh et al 2012) embedding dimension and a delay equal to two and four samples, respectively, were found for creating normalized ECG RPS. Then, the RPS was divided into small pixels (20 × 20 square grids).…”

“…The number of points in each square was normalized to the total number of points and the normalized numbers were considered as features (400 features). (e) Poincare section from ECG (13 features): reconstructed RPS explained in feature group (d) were used to extract 13 different features from an intersection of the ECG trajectory in RPS with a unity line (Poincare section of ECG (Parvaneh et al 2012(Parvaneh et al , 2016). Details on features and their estimation can be found in the original article (Parvaneh et al 2012).…”

Electrocardiogram Monitoring and Interpretation: From Traditional Machine Learning to Deep Learning, and Their Combination

“…As an example, extracted features from reconstructed phase space of ECG were proposed to capture changes in morphology of ECG due to AF through comparison of features between AF and NSR [13,14].…”

Shapelet Discovery for Atrial Fibrillation Detection

“…Furthermore, AF incidence and prevalence are likely to increase, especially with the aging of the population. ECG-based AF detectors mostly analyze atrial activities (e.g., absence of P waves) [2][3][4], ventricular responses (e.g., irregularity in RR intervals) [5][6][7], or both. Different features from RR intervals, such as heart rate variability (HRV), geometric representation, and entropy are used to differentiate between AF and Normal Sinus Rhythm (NSR) [7,8].…”