Pattern discrimination of neuroelectric waveforms evoked by synchronized visuo-motor integration

Today, there is a significant demand for fast, accurate, and automated methods for the discrimination of latent patterns in neuroelectric waveforms. One of the main challenges is the development of efficient feature extraction tools to utilize the rich spatio-temporal information inherent in large scale human electrocortical activity. In this paper, our aim is to isolate the most suitable feature extraction method for accurate classification of EEG data related to distinct modes of sensorimotor integration. Our results demonstrate that with some user-dependent input for feature space constraint, a simple classification framework can be developed to accurately distinguish between brain electrical activity patterns during two distinct conditions.