On the use of EEG features towards person identification via neural networks

Abstract

Person identification based on spectral information extracted from the EEG is addressed in this work a problem that has not yet been seen in a signal processing framework. Spectral features are extracted non-parametrically from real EEG data recorded from healthy individuals. Neural network classification is applied on these features using a Learning Vector Quantizer in an attempt to experimentally investigate the connection between a person's EEG and genetically specific information. The proposed method, compared with previously proposed methods, has yielded encouraging correct classification scores in the range of 80% to 100% (case-dependent). These results are in agreement with previous research showing evidence that the EEG carries genetic information.

Intelligent Training Algorithm for Artificial Neural Network EEG Classifications

“…Multi-layer Perceptron (MLP) neural network has been implemented to classify EEG signal by many researchers [17][18][19]. MLP networks are the most popular feed forward supervised artificial neural networks that map set of input data to a set of appropriate output.…”

Electrophysiological Biometrics: Opportunities and Risks

“…In fact, it is estimated that 2-3% of the population is missing the feature that is required for the authentication, or that the provided biometric sample is of poor quality. It has been proven that the EEG and ECG are unique enough to be used for biometric purposes (Marcel and Millán 2007 ;Mohammadi et al 2006 ;Paranjape et al 2001 ;Poulos et al 1998Poulos et al , 1999Poulos et al , 2001Poulos et al , 2002Riera et al 2008a, b ;Biel et al 2001 ;Chang 2005 ;Israel et al 2005 ;Kyoso 2001 ;Palaniappan and Krishnan 2004 ) . In fact, if we think on the huge number of neurons present in a typical adult brain (10^11) and their number of connections (10^15), we can defi nitively claim that no 2 brain are identical.…”

“…At any rate, the reader will agree with the statement that "the ultimate seat of identity lies in the living, dynamic brain", or, at least, in part of it (e.g., in the abstract set of neuronal connections). In recent work we have advanced a great deal in the development of physiologically based biometric systems exploiting EEG (Marcel and Millán 2007 ;Mohammadi et al 2006 ;Paranjape et al 2001 ;Poulos et al 1998Poulos et al , 1999Poulos et al , 2001Poulos et al , 2002Riera et al 2008a, b ) and ECG (Biel et al 2001 ;Chang 2005 ;Israel et al 2005 ;Kyoso 2001 ;Palaniappan and Krishnan 2004 ) signals and classifi cation algorithms. The derived systems rely on spontaneously generated electrophysiological signals, and as such they are in some sense weaker to spoofi ng attacks than they could be.…”

Multimodal Physiological Biometrics Authentication

“…This kind of multi-stage fusion architecture has been presented in [22] as an advancement for biometry systems.This paper describes a ready-to-use authentication biometric system based on EEG and ECG. This constitutes the first difference with already presented works [4,5,7,8,9,14,15,16,17,18,25]. The system presented herein undertakes subject authentication, whereas a biometric identification has been the target of those works.…”

“…A reduced number of electrodes have been already used in past works [4,5,7,8,9,25] in order to reduce system obtrusiveness. This feature has been implemented in our system.…”