P-Wave Configuration in the Signal Averaged Electrocardiogram: Which Filter Technique Differentiates Best Between Patients With Paroxysmal Atrial Fibrillation and Healthy Volunteers?

Early identification of patients at risk for paroxysmal atrial fibrillation (PAF) is essential to attain optimal treatment and a favorable prognosis. We compared the performance of a beat-to-beat (B2B) P-wave analysis with that of standard P-wave indices (SPWIs) in identifying patients prone to PAF. To this end, 12-lead ECG and 10 min vectorcardiogram (VCG) recordings were obtained from 33 consecutive, antiarrhythmic therapy naïve patients, with a short history of low burden PAF, and from 56 age- and sex-matched individuals with no AF history. For both groups, SPWIs were calculated, while the VCG recordings were analyzed on a B2B basis, and the P-waves were classified to a primary or secondary morphology. Wavelet transform was used to further analyze P-wave signals of main morphology. Univariate analysis revealed that none of the SPWIs performed acceptably in PAF detection, while five B2B features reached an AUC above 0.7. Moreover, multivariate logistic regression analysis was used to develop two classifiers—one based on B2B analysis derived features and one using only SPWIs. The B2B classifier was found to be superior to SPWIs classifier; B2B AUC: 0.849 (0.754–0.917) vs. SPWIs AUC: 0.721 (0.613–0.813), p value: 0.041. Therefore, in the studied population, the proposed B2B P-wave analysis outperforms SPWIs in detecting patients with PAF while in sinus rhythm. This can be used in further clinical trials regarding the prognosis of such patients.

Download Full-text

Methodology for detection of paroxysmal atrial fibrillation based on P-Wave, HRV and QR electrical alternans features

International Journal of Electrical and Computer Engineering (IJECE) ◽

10.11591/ijece.v10i4.pp4023-4034 ◽

2020 ◽

Vol 10 (4) ◽

pp. 4023

Author(s):

Henry Castro ◽

Juan David Garcia-Racines ◽

Alvaro Bernal-Noreña

Keyword(s):

Neural Network ◽

Atrial Fibrillation ◽

Paroxysmal Atrial Fibrillation ◽

P Wave ◽

Computational Techniques ◽

K Nearest Neighbors ◽

Ecg Signals ◽

Complex Process ◽

Artificial Neural Network Ann ◽

Electrical Alternans

The detection of Paroxysmal Atrial Fibrillation (PAF) is a fairly complex process performed manually by cardiologists or electrophysiologists by reading an electrocardiogram (ECG). Currently, computational techniques for automatic detection based on fast Fourier transform (FFT), Bayes optimal classifier (BOC), k-nearest neighbors (K-NNs), and artificial neural network (ANN) have been proposed. In this study, six features were obtained based on the morphology of the P-Wave, the QRS complex and the heart rate variability (HRV) of the ECG. The performance of this methodology was validated using clinical ECG signals from the Physionet arrhythmia database MIT-BIH. A feedforward neural network was used to detect the presence of PAF reaching a general accuracy of 97.4%. The results obtained show that the inclusion of the information of the P-Wave, HRV and QR Electrical alternans increases the accuracy to identify the PAF event compared to other works that use the information of only one or at most two of them.

Download Full-text