An Improved QRS Detection Approach based on Two-Lead Bi-Referring

Honghai Zhu, Xiaojuan Hu, and Jun Dong


QRS Detection, Multi-lead ECG, R Amplitude, Noise Amplitude, Signal to Noise Ratio


For improving QRS detection accuracy of single lead algorithms in multi-lead electrocardiogram (ECG), based on two-lead bi-referring, a new approach is introduced. As we know, there is much bursting noise in ECG and when it occurs in the detecting lead, it may decrease the accuracy of single lead QRS detecting algorithms. Meanwhile, in clinic, most of ECG is multi-lead, for example the famous MIT-BIH ECG database (MIT-BIH) is two leads system. So, in multi-lead system, it needs to be investigated how to improve existing QRS detection algorithms which are designed only with one lead ECG. In this paper, the final QRS positions are selected from the detection results of the cleanest lead which has least bursting noise and most regular r-peaks. It had been tested in MIT-BIH and Chinese cardiovascular disease database (CCDD) with four single lead algorithms and the results showed that it could improve their accuracy.

Important Links:

Go Back