QRS Complex Detection Using Zero Frequency Filtering

QRS complexes of electro cardiogram (ECG) signal are impulse-like in nature, which are considered as reference points in ECG analysis. Motivated by the zero frequency filter (ZFF) based epoch extraction from speech signals, in the present work, the application of ZFF for the QRS detection is exploited. ECG is preprocessed by moving average subtraction method to remove the base line drift and Hilbert envelope is computed. The preprocessed ECG is passed through a cascade of two resonators located at zero frequency and the trend of the resultant signal is removed by local average subtraction method. The proposed method is based only on the presence of impulse information at zero Hz and does not require any kind of decision logic or window for the selection of QRS. The proposed method is validated over MIT-BIH database using standard evaluation techniques and the results are compared with that of existing high performance stateof- the-art techniques.