ABSTRACT
Objective To develop a real-time QRS complex detection algorithm of dynamic ECG signals for the GPRS mobile telemonitoring system. Methods Before the first and second derivatives of ambulatory ECG signals were processed by moving average method, the signals sampled from CM5 monitoring lead were filtered with the average of continuous four ECG sample signal points. The R waves could be detected precisely by local minima of second derivatives and Q & S waves were located correctly by cross-zero points of first derivatives of ambulatory ECG signals in a short-time searching windows. The QRS recognition thresholds, which could revise themselves according to the detected values and vary with the analyzing signals, were designed in this paper. Results With a polynomial computation complexity, the novel algorithm insensitive to baseline draft and noise caused by mobile communication filtered power-line interference and most of muscle noise and reduced the search time below 0.02 s during detecting each Q wave, R wave and S wave. For the normal and clinical patients, this algorithm correctly detected up to 99.8% of the QRS complex of ambulatory ECG signals. Conclusion The algorithm can meet the need of real-time QRS complex detection and analysis for the GRRS mobile ECG telemonitoring system.