Comparing RR-Interval-Based and Whole-Signal-Based Machine Learning Models for Atrial Fibrillation Detection from Single-lead Electrocardiograms

The aim of this study was to compare the performance of machine learning models to detect atrial fibrillation (AF) from single-lead ECGs which use either RR-intervals alone, or the entire ECG signal. Experiments were conducted using single-lead, 30-second ECG signals acquired using handheld ECG recorders from two datasets: the Computing in Cardiology (CinC) 2017 dataset (public), and the Screening for Atrial Fibrillation with ECG to Reduce Stroke (SAFER) dataset (private). The models assessed in this study were: two models which used the whole ECG signal, both of which were top-performing models from the 2017 PhysioNet / CinC Challenge; and two RR-interval-based models - a state-of-the-art model and a novel model which detects AF from a 2D representation of the differences between RR intervals. The models had AUROCs of 0.93 - 0.99. The AUPRCs varied more widely, from 0.64-0.94. The novel RR-interval-based AF detection model achieved an AUPRC of 0.94 on the CinC 2017 dataset, outperforming the state-of-the-art RR-interval-based model (0.88) and the entire-signal-based models (0.68 and 0.64). This experiment demonstrated that AF detection models utilizing only RR intervals could achieve comparable performance to those utilizing the entire ECG signal.