A Multilead Fusion Based QRS Complex Detection Method on 12-lead Electrocardiogram Signals

The electrocardiogram (ECG) signal records the electrical activity of the heartbeat on one or more leads. The QRS complex is thought to be one of the most prominent ECG waves due to its large slope and high amplitude, and the accurate detection of the QRS complex is an essential function for the ECG analysis algorithms. Although numerous QRS complex detectors have been proposed, few method payed attention to analyzing the information simultaneous present in multilead ECG signals, which would be resistant to the noise. In this work, we presented a multilead QRS complex detection approach on 12-lead ECG signals. After detecting the QRS complex on each single lead, a novel multilead fusion strategy was applied to cluster the QRS complexes. The proposed method was evaluated on the St. Petersburg Institute of Cardiological Technics 12-lead Arrhythmia Database. The detection error rate (DER) of QRS complex ranged from 1.17% to 13.78% on each single lead. After the multilead fusion, the DER was reduced to 0.39%. Experimental results show that the proposed method would be useful to produce an accurate QRS complex detection result.