Removal of ocular artifacts from EEG using an efficient neuralnetwork based adaptive filtering technique

摘要

The electroencephalogram (EEG) is susceptible to various largensignal contaminations or artifacts. Ocular artifacts act as major sourcenof noise, making it difficult to distinguish normal brain activitiesnfrom the abnormal ones. In this letter, an efficient technique thatncombines two popular adaptive filtering techniques, namely adaptivennoise cancellation and adaptive signal enhancement, in a singlenrecurrent neural network is proposed for the adaptive removal of ocularnartifacts from EEG. A real time recurrent learning algorithm is employednfor training the proposed neural network which converges faster to anlower mean squared error. This technique is suitable for real-timenprocessing