Removing Artifacts From Eeg Signal Using Wavelet Transform and Conventional Filters

Abstract

Electroencephalogram (EEG) is a signal of an electrical nature reflecting the neuronal activities of the brain. It is used for the diagnosis of certain cerebral pathologies. However, it becomes more difficult to identify and analyze it when it is corrupted by artifacts of non-cerebral origin such as eye movements, cardiac activities ..., therefore, it is essential to remove these parasitic signals. In literature, there are different techniques for removing artifacts. This paper proposes and discusses a new EEG de-noising technique, based on a combination of wavelet transforms and conventional filters. The results of the proposed method are evaluated using three common criteria: signal-to-noise-ratio (SNR), mean square error (MSE) and cross correletion function (CCF). These experimental results demonstrate that the proposed approach can be an effective tool for removing artifact without suppression of any signal components.