Classification of P300 Component Using a Riemannian Ensemble Approach

Dominik Krzemiński; Sebastian Michelmann; Matthias Treder; Lorena Santamaria

doi:10.1007/978-3-030-31635-8_229

Classification of P300 Component Using a Riemannian Ensemble Approach

Dominik Krzemiński, Sebastian Michelmann, Matthias Treder, Lorena Santamaria

Psychology

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

We present a framework for P300 ERP classification on the 2019 IFMBE competition dataset using a combination of a Riemannian geometry and ensemble learning. Covariance matrices and ERP prototypes are extracted after the EEG is passed through a filter bank and an ensemble of LDA classifiers is trained on subsets of channels, trials, and frequencies. The model selects a final class based on maximum probability of evidence from all ensembles. Our pipeline achieves an average classification accuracy of 81.2% on the test set.

Original language	English (US)
Title of host publication	15th Mediterranean Conference on Medical and Biological Engineering and Computing – MEDICON 2019 - Proceedings of MEDICON 2019
Editors	Jorge Henriques, Paulo de Carvalho, Nuno Neves
Publisher	Springer
Pages	1885-1889
Number of pages	5
ISBN (Print)	9783030316341
DOIs	https://doi.org/10.1007/978-3-030-31635-8_229
State	Published - 2020
Event	15th Mediterranean Conference on Medical and Biological Engineering and Computing, MEDICON 2019 - Coimbra, Portugal Duration: Sep 26 2019 → Sep 28 2019

Publication series

Name	IFMBE Proceedings
Volume	76
ISSN (Print)	1680-0737
ISSN (Electronic)	1433-9277

Conference

Conference	15th Mediterranean Conference on Medical and Biological Engineering and Computing, MEDICON 2019
Country/Territory	Portugal
City	Coimbra
Period	9/26/19 → 9/28/19

Keywords

Brain-computer interface
Classification
ERP
Ensemble learning
LDA
Logistic regression
P300
Riemannian geometry

ASJC Scopus subject areas

Bioengineering
Biomedical Engineering

Access to Document

10.1007/978-3-030-31635-8_229

Cite this

Krzemiński, D., Michelmann, S., Treder, M., & Santamaria, L. (2020). Classification of P300 Component Using a Riemannian Ensemble Approach. In J. Henriques, P. de Carvalho, & N. Neves (Eds.), 15th Mediterranean Conference on Medical and Biological Engineering and Computing – MEDICON 2019 - Proceedings of MEDICON 2019 (pp. 1885-1889). (IFMBE Proceedings; Vol. 76). Springer. https://doi.org/10.1007/978-3-030-31635-8_229

Classification of P300 Component Using a Riemannian Ensemble Approach. / Krzemiński, Dominik; Michelmann, Sebastian; Treder, Matthias et al.
15th Mediterranean Conference on Medical and Biological Engineering and Computing – MEDICON 2019 - Proceedings of MEDICON 2019. ed. / Jorge Henriques; Paulo de Carvalho; Nuno Neves. Springer, 2020. p. 1885-1889 (IFMBE Proceedings; Vol. 76).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Krzemiński, D, Michelmann, S, Treder, M & Santamaria, L 2020, Classification of P300 Component Using a Riemannian Ensemble Approach. in J Henriques, P de Carvalho & N Neves (eds), 15th Mediterranean Conference on Medical and Biological Engineering and Computing – MEDICON 2019 - Proceedings of MEDICON 2019. IFMBE Proceedings, vol. 76, Springer, pp. 1885-1889, 15th Mediterranean Conference on Medical and Biological Engineering and Computing, MEDICON 2019, Coimbra, Portugal, 9/26/19. https://doi.org/10.1007/978-3-030-31635-8_229

Krzemiński D, Michelmann S, Treder M, Santamaria L. Classification of P300 Component Using a Riemannian Ensemble Approach. In Henriques J, de Carvalho P, Neves N, editors, 15th Mediterranean Conference on Medical and Biological Engineering and Computing – MEDICON 2019 - Proceedings of MEDICON 2019. Springer. 2020. p. 1885-1889. (IFMBE Proceedings). doi: 10.1007/978-3-030-31635-8_229

Krzemiński, Dominik ; Michelmann, Sebastian ; Treder, Matthias et al. / Classification of P300 Component Using a Riemannian Ensemble Approach. 15th Mediterranean Conference on Medical and Biological Engineering and Computing – MEDICON 2019 - Proceedings of MEDICON 2019. editor / Jorge Henriques ; Paulo de Carvalho ; Nuno Neves. Springer, 2020. pp. 1885-1889 (IFMBE Proceedings).

@inproceedings{49889cf9204d4d1fb4eadcc8b2a9c5da,

title = "Classification of P300 Component Using a Riemannian Ensemble Approach",

abstract = "We present a framework for P300 ERP classification on the 2019 IFMBE competition dataset using a combination of a Riemannian geometry and ensemble learning. Covariance matrices and ERP prototypes are extracted after the EEG is passed through a filter bank and an ensemble of LDA classifiers is trained on subsets of channels, trials, and frequencies. The model selects a final class based on maximum probability of evidence from all ensembles. Our pipeline achieves an average classification accuracy of 81.2% on the test set.",

keywords = "Brain-computer interface, Classification, ERP, Ensemble learning, LDA, Logistic regression, P300, Riemannian geometry",

author = "Dominik Krzemi{\'n}ski and Sebastian Michelmann and Matthias Treder and Lorena Santamaria",

note = "Publisher Copyright: {\textcopyright} 2020, Springer Nature Switzerland AG.; 15th Mediterranean Conference on Medical and Biological Engineering and Computing, MEDICON 2019 ; Conference date: 26-09-2019 Through 28-09-2019",

year = "2020",

doi = "10.1007/978-3-030-31635-8_229",

language = "English (US)",

isbn = "9783030316341",

series = "IFMBE Proceedings",

publisher = "Springer",

pages = "1885--1889",

editor = "Jorge Henriques and {de Carvalho}, Paulo and Nuno Neves",

booktitle = "15th Mediterranean Conference on Medical and Biological Engineering and Computing – MEDICON 2019 - Proceedings of MEDICON 2019",

}

TY - GEN

T1 - Classification of P300 Component Using a Riemannian Ensemble Approach

AU - Krzemiński, Dominik

AU - Michelmann, Sebastian

AU - Treder, Matthias

AU - Santamaria, Lorena

PY - 2020

Y1 - 2020

N2 - We present a framework for P300 ERP classification on the 2019 IFMBE competition dataset using a combination of a Riemannian geometry and ensemble learning. Covariance matrices and ERP prototypes are extracted after the EEG is passed through a filter bank and an ensemble of LDA classifiers is trained on subsets of channels, trials, and frequencies. The model selects a final class based on maximum probability of evidence from all ensembles. Our pipeline achieves an average classification accuracy of 81.2% on the test set.

AB - We present a framework for P300 ERP classification on the 2019 IFMBE competition dataset using a combination of a Riemannian geometry and ensemble learning. Covariance matrices and ERP prototypes are extracted after the EEG is passed through a filter bank and an ensemble of LDA classifiers is trained on subsets of channels, trials, and frequencies. The model selects a final class based on maximum probability of evidence from all ensembles. Our pipeline achieves an average classification accuracy of 81.2% on the test set.

KW - Brain-computer interface

KW - Classification

KW - ERP

KW - Ensemble learning

KW - LDA

KW - Logistic regression

KW - P300

KW - Riemannian geometry

UR - http://www.scopus.com/inward/record.url?scp=85075860202&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85075860202&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-31635-8_229

DO - 10.1007/978-3-030-31635-8_229

M3 - Conference contribution

AN - SCOPUS:85075860202

SN - 9783030316341

T3 - IFMBE Proceedings

SP - 1885

EP - 1889

BT - 15th Mediterranean Conference on Medical and Biological Engineering and Computing – MEDICON 2019 - Proceedings of MEDICON 2019

A2 - Henriques, Jorge

A2 - de Carvalho, Paulo

A2 - Neves, Nuno

PB - Springer

T2 - 15th Mediterranean Conference on Medical and Biological Engineering and Computing, MEDICON 2019

Y2 - 26 September 2019 through 28 September 2019

ER -

Classification of P300 Component Using a Riemannian Ensemble Approach

Abstract

Publication series

Conference

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this