On QRS detection methodologies: A revisit for mobile phone applications, wireless ECG monitoring and large ECG databases analysis

This review discusses the development of any QRS detection algorithm based on three assessment criteria: robustness to noise, parameter choice, and numerical efficiency. This is because of there is a need nowadays for fast and robust QRS detection algorithms to be deployed on mobile phones, and run over large ECG databases and wireless medical body area network. Till now, there is no satisfying universal solution for detecting QRS complex given the fact it was a topic of investigation for the last 30 years. The difficulty arises mainly because of the diversity of the QRS waveforms, abnormalities, low SNR and the artifacts accompanying the ECG signals. The existing QRS detection algorithms may provide an acceptable solution on small ECG segments, within a certain amplitude range, amid particular type of arrhythmia, or/and noise. These issues are discussed in comparison with the most conventional algorithms, followed by recommendations for developing a reliable QRS detection suitable for large-recorded ECG signals and battery-