Accurate QRS complex detection in 12-lead ECG signals using multi-lead fusion

Abstract

In this paper, a novel multi-lead fusion approach for accurate detection of the QRS complex from 12-lead ECG (Electrocardiogram) signals is proposed. The framework is divided into two major stages. In the first stage, the single-lead QRS complex detection algorithm detects the QRS complex location of each lead. In the second stage, the multi-lead fusion method combines the QRS complex locations obtained in each of the 12-leads, thereby improving the performance of QRS complex detection in terms of sensitivity and positive predictivity by discarding the false positives detected in each lead. The performance of the proposed method is validated using the databases of the St. Petersburg Institute of Cardiological Technics (INCART) (Goldberger et al., 2000) and the CSE (Common Standards for Electrocardiography) (Willems et al., 1987). The proposed framework achieves 99.87% sensitivity and 99.96% positive predictive accuracy for the INCART database. Sensitivity and positive predictivity for the CSE database were found to be 100% and 99.13%, respectively.