Atrial Fibrillation Screening through Combined Timing Features of Short Single-Lead Electrocardiograms

Abstract

Atrial fibrillation (AF) is the most common cardiac arrhythmia, as well as a growing healthcare burden worldwide. It is often asymptomatic and usually starts with very brief episodes, thus making its early detection an interesting challenge. For that purpose, the present work introduces a novel method exploiting the variability presented both by ventricular and atrial activities reflected on the surface electrocardiogram (ECG). Thus, time series from the RR intervals and the fibrillatory waves morphology contained by the TQ intervals are first generated and, then, their regularity is estimated making use of the Coefficient of Sample Entropy (COSEn). The collected information is finally combined through a multi-class support vector machine (SVM) approach to discern among short episodes of AF, normal sinus rhythm (NSR) and other rhythms (OR). The algorithm has been validated in the context of the PhysioNet Computing in Cardiology Challenge 2017, thus reporting a global F 1 measure of 0.73 for the training set and 0.71 for the testing group. Nonetheless, to evaluate the method in a common scenario for previous works, the widely used MIT-BIH AF database has also been considered. A notably higher F 1 score of 0.87 has been provided in this case. The significantly different balance between the number of AF, NSR and OR recordings in both databases could justify the obtained outcomes.

Enhanced Atrial Fibrillation Detection-based Wavelet Scattering Transform with Time Window Selection and Neural Network Integration

“…Our model, which combines WSN, ANN, Time window, and Majority vote technique, achieves the highest overall accuracy. When contrasted with the findings of Pandey et al [18], as presented in Table V…”

“…In this paper, a comparison was conducted with state-of-art models designed for AF detection. Notably, Garcia et al [18] introduced a method that utilizes surface ECG data, capturing variability in ventricular and atrial activities. The approach involves generating time series data from R_R intervals and morphological features of fibrillatory waves in T_Q intervals.…”

A low-complexity algorithm for detection of atrial fibrillation using an ECG

“…Teijeiro et al (2017)). Several entries designed models using combined timing features and support vector machines (García et al 2017) or convolutional neural networks (Schwab et al 2017, Sopic et al 2017, Zihlmann et al 2017. As noted by the competition organisers, the complex methods failed in gaining much advantage over simpler methods such as a standard random forest technique equipped by 'well chosen' features (Clifford et al 2017).…”

Automatic Detection of Atrial Fibrillation from ECG Signal Using Hybrid Deep Learning Techniques

“…Much work has gone into ECG categorization, and more is being done in the process. In [19], Garcia et al propose a new strategy that takes advantage of ventricular and atrial activity variability, as shown on the surface electrocardiogram (ECG). First, the time series generated from RR intervals and fibrillatory wave morphology derived from TQ intervals are developed.…”

“…The work demonstrated by Garcia et al [19] used a multiclass SVM approach for classification and achieved an F1 score of 73%. In comparison, Rajpurkar et al [20] used the approach of ResNet (34 layers) that converts the sequence of ECG samples into a sequence of rhythm classes.…”