Support system for classification of beat-to-beat arrhythmia based on variability and morphology of electrocardiogram

Abstract

Background: Several authors use the R-R interval, which is the temporal difference between the largest waves (R waves) of the electrocardiogram (ECG), to propose a support system for the diagnosis of arrhythmias. However, R-R interval analysis does not measure ECG waveform deformations such as P wave deformations for atrial fibrillation. Objective: In this study, we propose an arbitrary analysis the any segment of the heartbeat. This analysis is a generalization of a previous work that measures the wave deformations of the ECG signal. Methods: We proposed to investigate the voltage (mV) variation occurring at each heartbeat interval using statistical moments. Unlike the R-R interval in which each heartbeat is associated with a single real number, the proposed method associates each heartbeat to a set of points, that is, a vector. The heartbeats were obtained in the following databases: MIT-BIH Normal Sinus Rhythm, MIT-BIH Atrial Fibrillation (AF), and MIT-BIH Arrhythmia; and the classifiers used to evaluate the proposed method were linear discriminant analysis, k-nearest neighbors, and support vector machine. The experiments were conducted using 80% of the patients for training (16 healthy patients, 41 patients with arrhythmia, and 20 patients with AF) and 20% of the patients for testing (2 healthy patients, 6 patients with arrhythmia, and 3 patients with AF). Results: The proposed method proved to be efficient in solving global (accuracy is up to 99.78% in the arrhythmia classification) and local (accuracy of 100% in the AF classification) heartbeat problems. Conclusion: The results obtained by the proposed method can be used to support decision-making in clinical practices.

Full Issue (Vol. 10, December 2022)

“…The confusion matrix of the proposed CNN model is shown in Figure 6. Here, three statistical parameters have been computed to evaluate the performance of our classifier which are as follows (Queiroz et al, 2019):…”

“…The arrhythmias are most serious among them, and may cause sudden cardiac arrest or stroke (Huikuri, Castellanos, & Myerburg, 2001) (Chen, Wang, & Wang, 2018;De Chazal & Reilly, 2006;Mitra, Mitra, & Chaudhuri, 2006) have been extracted from the time-domain analysis . Moreover, statistical features have been extracted in terms of variance, kurtosis, and skewness (Queiroz, Azoubel, & Barros, 2019;Queiroz, Junior, Lucena, & Barros, 2018). Besides, many literatures implemented wavelet transform to classify ECG arrhythmias in timefrequency analysis (Banerjee & Mitra, 2013;Khorrami & Moavenian, 2010).…”

A Novel Approach to Detect Cardiac Arrhythmia Based on Continuous Wavelet Transform and Convolutional Neural Network

“…The confusion matrix of the proposed CNN model is shown in Figure 6. Here, three statistical parameters have been computed to evaluate the performance of our classifier which are as follows (Queiroz et al, 2019):…”

“…Here, the features like heart rate, RR interval, R amplitude, PR interval, QRS duration, P-wave, and T-wave duration (Chen, Wang, & Wang, 2018;De Chazal & Reilly, 2006;Mitra, Mitra, & Chaudhuri, 2006) have been extracted from the time-domain analysis . Moreover, statistical features have been extracted in terms of variance, kurtosis, and skewness (Queiroz, Azoubel, & Barros, 2019;Queiroz, Junior, Lucena, & Barros, 2018). Besides, many literatures implemented wavelet transform to classify ECG arrhythmias in timefrequency analysis (Banerjee & Mitra, 2013;Khorrami & Moavenian, 2010).…”