Removal of ECG Artifacts from EEG Signals

Abstract

Any unusual feature in brain functioning, structure, or biochemical levels is referred to as brain abnormality. Brain abnormalities, deformities, or dysfunction will affect the whole body. Electroencephalogram (EEG) tests are taken in order to diagnose many diseases caused by brain abnormalities such as sleep disorders, head injuries, Alzheimer’s disease, Epilepsy, brain hemorrhage and etc. However, an EEG recording could have many types of noises or artifacts that came from the blinking of the eye, heartbeat, muscle movement, and many other types of noises which will contaminate the EEG recording, decreasing the accuracy of the EEG recording. In this paper, a machine learning algorithm was proposed and used to remove electrocardiogram (ECG) artifacts from the EEG signal. ECG artifacts are noises that came from the beating of the heart. The Independent Component Analysis (ICA) algorithm was the machine learning algorithm that was used for artifact removal. It was implemented by using Python code and was executed in Google Colaboratory. A completely clean EEG signal free from any artifacts is impossible to obtain but implementing this machine learning algorithm to remove ECG artifacts from the EEG signal, will produce a better EEG signal