Applying Auto Regression Techniques on Amyotrophic Lateral Sclerosis Patients EEG Dataset with P300 Speller

This paper deals with the application of auto regression techniques to find the best fitting curves for the Electroencephalograph (EEG) data of Amyotrophic Lateral Sclerosis (ALS) patients with P300 speller. ALS is a degenerative neuron disease bringing gradual impairment of motor neurons leading to total loss of voluntary limb movement in sometime. A P300 speller is a 6X6 matrix of English alphabets in which each column and each row is highlighted periodically and the patient has to concentrate on the correct alphabet to evoke P300 event related potential. Auto regression is a curve fitting technique for sampled data. The best fit obtained in this study for the ALS patients’ EEG channels which can used to predict incomplete or subsequent EEG data to enhance communication through P300 speller.