Low-distortion baseline removal algorithm for electrocardiogram signals

A new ECG baseline removal method is presented in this paper. This method is based on the Savitzky-Golay polynomial smoothing filter. Since this filter is defined in the time domain, it has the advantage of following the trend of the baseline wander. The results from this method are compared to the cubic spline method and a heart rate adaptive high-pass filter. The comparison shows that the SG baseline filter has comparable performance to these two other established methods. This new method is simple, and does not require extra knowledge about the ECG, such as isoelectric points required by the cubic spline method or the heart rate required by the high-pass filter. Thus it is more suitable for implementation in ECG monitoring systems.