Multiscale entropy analysis of attention ralated EEG based on motor imaginary potential

Abstract

In China, there are approximate 1.3% to 13.4% of children who have Attention Deficit Hyperactivity Disorder (ADHD), which may affect their physiology and psychology development badly. Attention related electroencephalograph (EEG) signals during the limbs motor imagery can be used to tell the different levels of people's attention. Such an EEG-based attention level discrimination can provide a method in curing ADHD and it can also be used in curing Altheimer's Disease patients. The conventional methods purpose the feature extraction of limbs motor imagery. In this study, Multiscale Entropy (MSE) is introduced to discriminate the EEG signals recorded during three attention tasks. We have discriminated the different attention states by using this method, with 63.158% accuracy to some subjects. The effectiveness of the method is proved by our experiment.