The electrocardiogram -ECG- describes the electrical activity of the heart. It is obtained by placing electrodes on the chest, arms and legs. With every heartbeat, an impulse travels through the heart, which determines its rhythm and rate and causes the heart muscle to contract and pump blood. The voltage variations measured by the electrodes are cause by the action potentials of the excitable cardiac cells, as they make the cells contract. The ECG is characterized by a series of waves whose morphology and timing provide information used for diagnosing diseases reflected by disturbances of the electrical activity of the heart. The time pattern that characterizes the occurrence of successive heartbeats is also very important.
The first ECG recording device was developed by the Dutch physiologist Willem Einthoven, using a string galvanometer which was sensitive enough to record electrical potentials on the body surface. He also defined sites for electrode placement on the arms and legs which remain in use today. Since then, ECG recording has developed incredibly and become an indispensable tool in many different contexts. The ECG record is used today in a widevariety of clinical applications. Its importance has been strengthened thanks to the discoveries of subtle variability patterns which are present in rhythm or wave morphology.
The electrodes used for ECG recording are positioned so that the spatiotemporal variations of the cardiac electrical field are sufficiently well-reflected. The difference in voltage between a pair of electrodes is referred to as a lead. The ECG is typically recorded with a multiple-lead configuration. The electrode wires are connected to a differential amplifier specially designed for bioelectrical signals. The ECG ranges from a few microvolts to about 1V in magnitude. Whereas the characteristic waves of an ECG have a maximal magnitude of only few millivolts, a wandering baseline in the ECG due to variations in electrode-skin impedance may reach 1V. The amplifier bandwith is commonly between 0.05 and 100-500Hz. The characteristic waves of an ECG are shown in the following image.Atrial depolarization is reflected by the P wave, and ventricular depolarization is reflected by the QRS complex, whereas the T wave reflects ventricular repolarization. The amplitude of a wave is measured with reference to the ECG baseline level, commonly defined by the isoelectric line which immediately precedes the QRS complex.
There are several most common types of noise and artifacts in the ECG. The baseline wander is an extraneous, low-frequency activity in the ECG which may interfere with the signal analysis, making the clinical interpretation inaccurate. When baseline wander takes place, ECG measurements related to the isoelectric line cannot be computed since it is not well-defined. Baseline wander is often exercise-induced and may have its origin in a variety of sources, including perspiration, respiration, body movements and poor electrode contact. The spectral content of the baseline wander is usually in the range between 0.05-1Hz but, during strenuous exercise, it may contain higher frequencies.
Generally, methods used to reduce this kind of disturbance can be divided into two groups: methods based on baseline wander estimation and methods based on high-pass filtering. In the first case, baseline wander is usually estimated using a polynomial and next this estimate is subtracted from the disturbed signal. For this approach the determination of characteristic points (knots) is necessary. These methods are very sensitive to the knots determination errors and a catastrophic deterioration of their performance for significant ones may easily occur. The greatest disadvantage of the methods based on a high-pass filtering is a signal distortion due to the overlapping of signal and disturbance spectra. In other words, it is impossible to reduce the baseline wander without signal distortion using linear filtering. Additionally, due to the specific form of ECG signal, i.e. impulse-like nature, non-stationarity, non-symmetry with respect to the baseline, linear high-pass filters designed by the traditional methods introduce unacceptable signal distortion. If we take into account that (i) impulse-like part of ECG signal (QRS complexes) have a short duration, (ii) a baseline wander is a low-frequency disturbance, which is not synchronized with QRS complexes, then it is obvious that the best estimate of baseline wander is a sample mean in the moving window. Thus, a very simple and frequently used method for baseline wander reduction is the high-pass filtering based on a moving average (MA) filter. A high-pass filter to baseline wander reduction may also be realized as a one-weight adaptive filter. A new method presented in this paper uses a nonlinear modification of the moving average filter.