Analysing Eeg Sub-bands to Distinguish between Individuals with Neural Disorder Using Back Propagation Neural Network

The electroencephalogram (EEG) signal plays a vital role in the detection of various types of neural disorders such as autism spectrum disorder, epilepsy, Alzheimer’s, etc. Since EEG signals are lengthy and it contains a vast amount of useful data and artifacts analysing these signals by an expert using traditional methods are monotonous and time consuming. Recent years, many automatic diagnostic systems for examining EEG signals do exist. This paper proposes an automated system for analysing EEG signals using artificial neural networks to distinguish autism spectrum disorder subjects with normal subjects. We considered various statistical features of EEG signals as input for the proposed neural network model. The system is trained using back propagation neural network algorithm. The proposed method shows overall accuracy values as high as 88.8% can be achieved.