Removal of baseline drift from ECG signals based on morphology and adaptive filter

Baseline drift has a major impact on the preprocessing and feature extraction of the ECG signal. This paper analyzes the characteristics of baseline drift of Holter ECG signals. A novel method based on mathematical morphology combined with adaptive filter is proposed to eliminate the baseline drift in Holter ECG. The method uses morphological method to remove high frequency noise and extract the baseline drift. The baseline drift is subsequently used as the reference input of an adaptive filter for noise cancellation. Experiments show that the method is capable of effectively removing the ECG baseline drift with simple computation.