A neural network-based method using spectral EEG features achieved correct classification scores ranging from 80% to 100% for person identification in healthy individuals.
A direct connection between the electroencephalogram (EEG) and the genetic information of an individual has been suspected and investigated by neurophysiologists and psychiatrists since 1960. However, most of this early as well as more recent research focuses on the classification of pathological EEG cases, aiming to construct tests for purposes of diagnosis. On the contrary, our work focuses on healthy individuals and aims to establish an one-to-one correspondence between the genetic information of the individual and certain features of his/her EEG, as an intermediate step towards the further goal of developing a test for person identification based on features extracted from the EEG. Potential applications include, among others, information encoding and decoding and access to secure information. At the present stage the proposed method uses spectral information extracted from the EEG non-parametrically via the FFT and employs a neural network (a learning vector quantizer-LVQ) to classify unknown EEGs as belonging to one of a finite number of individuals. Correct classification scores ranging from 80% to 100% in experiments conducted on real data, show evidence that the EEG indeed carries genetic information and that the proposed method can be used to construct person identification tests based on EEG features.
Poulos et al. (Fri,) conducted a other in Healthy individuals. Neural network (learning vector quantizer-LVQ) using spectral EEG features was evaluated on Correct classification scores for person identification. A neural network-based method using spectral EEG features achieved correct classification scores ranging from 80% to 100% for person identification in healthy individuals.