Show simple item record

dc.contributor.authorSchetinin, Vitalyen
dc.contributor.authorJakaite, Livijaen
dc.contributor.authorNyah, Ndifrekeen
dc.contributor.authorNovakovic, Dusicaen
dc.contributor.authorKrzanowski, Wojteken
dc.date.accessioned2018-02-19T10:46:27Z
dc.date.available2018-02-19T10:46:27Z
dc.date.issued2018-01-26
dc.identifier.citationSchetinin V, Jakaite L, Nyah N, Novakovic D, Krzanowski W (2018) 'Feature extraction with GMDH-type neural networks for EEG-based person identification', International Journal of Neural Systems, 28 (6) 1750064en
dc.identifier.issn0129-0657
dc.identifier.pmid29370728
dc.identifier.doi10.1142/S0129065717500642
dc.identifier.urihttp://hdl.handle.net/10547/622506
dc.description.abstractThe brain activity observed on EEG electrodes is influenced by volume conduction and functional connectivity of a person performing a task. When the task is a biometric test the EEG signals represent the unique “brain print”, which is defined by the functional connectivity that is represented by the interactions between electrodes, whilst the conduction components cause trivial correlations. Orthogonalization using autoregressive modeling minimizes the conduction components, and then the residuals are related to features correlated with the functional connectivity. However, the orthogonalization can be unreliable for high-dimensional EEG data. We have found that the dimensionality can be significantly reduced if the baselines required for estimating the residuals can be modeled by using relevant electrodes. In our approach, the required models are learnt by a Group Method of Data Handling (GMDH) algorithm which we have made capable of discovering reliable models from multidimensional EEG data. In our experiments on the EEG-MMI benchmark data which include 109 participants, the proposed method has correctly identified all the subjects and provided a statistically significant (p<0.01) improvement of the identification accuracy. The experiments have shown that the proposed GMDH method can learn new features from multi-electrode EEG data, which are capable to improve the accuracy of biometric identification.
dc.description.sponsorshipThe research has been partly supported by the UK Leverhulme Trust, Grant F/00 811/Aen
dc.language.isoenen
dc.publisherWorld Scientific Journalsen
dc.relation.urlhttp://www.worldscientific.com/doi/abs/10.1142/S0129065717500642en
dc.rightsYellow - can archive pre-print (ie pre-refereeing)
dc.rights.urihttp://creativecommons.org/licenses/by-nc-nd/4.0/*
dc.subjectbiometricsen
dc.subjectbrain functional connectivityen
dc.subjectvolume conductionen
dc.subjectfeature extractionen
dc.subjectgroup method of data handlingen
dc.subjectC191 Biometryen
dc.titleFeature extraction with GMDH-type neural networks for EEG-based person identificationen
dc.typeArticleen
dc.contributor.departmentUniversity of Bedfordshireen
dc.contributor.departmentUniversity of Exeteren
dc.identifier.journalInternational Journal of Neural Systemsen
dc.date.updated2018-02-19T10:32:57Z
dc.description.notefile is post-print in the publisher's template (not the final version), confirmed by Vitaly by email 19/2/18
html.description.abstractThe brain activity observed on EEG electrodes is influenced by volume conduction and functional connectivity of a person performing a task. When the task is a biometric test the EEG signals represent the unique “brain print”, which is defined by the functional connectivity that is represented by the interactions between electrodes, whilst the conduction components cause trivial correlations. Orthogonalization using autoregressive modeling minimizes the conduction components, and then the residuals are related to features correlated with the functional connectivity. However, the orthogonalization can be unreliable for high-dimensional EEG data. We have found that the dimensionality can be significantly reduced if the baselines required for estimating the residuals can be modeled by using relevant electrodes. In our approach, the required models are learnt by a Group Method of Data Handling (GMDH) algorithm which we have made capable of discovering reliable models from multidimensional EEG data. In our experiments on the EEG-MMI benchmark data which include 109 participants, the proposed method has correctly identified all the subjects and provided a statistically significant (p<0.01) improvement of the identification accuracy. The experiments have shown that the proposed GMDH method can learn new features from multi-electrode EEG data, which are capable to improve the accuracy of biometric identification.


Files in this item

Thumbnail
Name:
main_R2.pdf
Size:
1.546Mb
Format:
PDF
Description:
author's version

This item appears in the following Collection(s)

Show simple item record

Yellow - can archive pre-print (ie pre-refereeing)
Except where otherwise noted, this item's license is described as Yellow - can archive pre-print (ie pre-refereeing)