Baseline Wandering Removal from Human Electrocardiogram Signal using Projection Pursuit Gradient Ascent Algorithm

Baseline noise removal from electrocardiogram (ECG) signal is a blind source separation problem. Various noises affect the measured ECG signal. Major ECG noises are baseline noise, electrode contact noise, muscle noise, instrument noise. Baseline noise distorts the low frequency segment of ECG signal. The low frequency segment is s-t segment. This segment is very important and has the information related to heart attack. People apply various algorithms to remove this noise from noisy ECG signal. We have applied projection pursuit gradient ascent algorithm to remove this noise from the measured ECG signal. This algorithm separates the independent signals from a mixture of signals. Efficient removal of baseline noise might give us certain information that are hidden from the doctors until now which may save the life of a person. Results for different baseline noise signals were analyzed. Different signal from MIT-BIH database were also analyzed for error in term of standard deviation and mean of error signal. Finally we did a comparative study of the results of different algorithms like kalman filter, cubic spline and moving average algorithms and showed that projection pursuit is the efficient one. Index Term-Baseline Noise, Cubic spline, Electrocardiogram, heart attack, Kalman filter, projection pursuit.