This chapter introduces different resources about noise in heart signals. It also provides a short explanation about artificial neural network (ANN), particle swarm optimization (PSO), and presents some of the previous studies related to heart signal noise removal, intelligent methods for detection of disorders, and feature extraction.
Top6.2 Noise In Ecg Signal
There are various sources of noise including imperfect contact of electrodes to the body, machine malfunction, electrical noise from elsewhere in the body, respiration and muscle contractions, which corrupts the ECG signal (Poungponsri and Yu, 2009). The produced noise consists of low-frequency components and high-frequency components that cause baseline wander and powerline interference, respectively (Jesmin et al., 2011). Below is a description for different types of noise in ECG signals.
6.2.1 Powerline Interference
Powerline interference is the 50 or 60 Hz pickup that is caused by improper grounding (Behbahani, 2007). This is due to an impulse or spike at 60 Hz/50 Hz harmonics which can be removed using a 60 Hz notch filter. Figure 1 shows the powerline interference that has corrupted the ECG.
Figure 1. 60 Hz powerline interference
Another type of noise that is caused by poor contact between the electrodes and skin is transient interference that effectively reduces the accuracy of the measurement device. Figure 2 shows the false beat in the signal, which is because of the imperfect electrode contact. The contact may be loose permanently, or intermittently. This loose contact may be as the result of the movement and vibration of the electrodes on the skin (Pomi and Olivera, 2006) and can result in large artefacts.
6.2.3 Motion Artefact
Motion artefacts are transient baseline change due to the electrode skin impedance with electrode motion. This motion generates the larger amplitude in the ECG signal. The motion artefact sometimes has the peak amplitude of 500 percent of the ECG. Its duration is about 100 – 500 ms (Pomi and Olivera, 2006). This noise can be removed using an adaptive filter. Figure 3 shows the motion artefact that cause similar amplitude for P, QRS and T in waveform.
Skin stretching causes changing in the amplitude of P, QRS, and T wave, which is because of some mechanical disturbances of the distribution of electrode-electrolyte interface. It causes higher amplitude for P, QRS, and T wave than the normal.