Robust common spatial filters with a maxmin approach

Motoaki Kawanabe, Carmen Vidaurre, Simon Scholler, Klaus Muller

Research output: Chapter in Book/Report/Conference proceedingConference contribution

14 Citations (Scopus)

Abstract

Electroencephalographic signals are known to be non-stationary and easily affected by artifacts, therefore their analysis requires methods that can deal with noise. In this work we present two ways of calculating robust common spatial patterns under a maxmin approach. The worst-case objective function is optimized within prefixed sets of the covariance matrices that are defined either very simply as identity matrices or in a data driven way using PCA. We test common spatial filters derived with these two approaches with real world brain-computer interface (BCI) data sets in which we expect substantial "day-to-day" fluctuations (session transfer problem). We compare our results with the classical common spatial filters and show that both can improve the performance of the latter.

Original languageEnglish
Title of host publicationProceedings of the 31st Annual International Conference of the IEEE Engineering in Medicine and Biology Society: Engineering the Future of Biomedicine, EMBC 2009
Pages2470-2473
Number of pages4
DOIs
Publication statusPublished - 2009 Dec 1
Externally publishedYes
Event31st Annual International Conference of the IEEE Engineering in Medicine and Biology Society: Engineering the Future of Biomedicine, EMBC 2009 - Minneapolis, MN, United States
Duration: 2009 Sep 22009 Sep 6

Other

Other31st Annual International Conference of the IEEE Engineering in Medicine and Biology Society: Engineering the Future of Biomedicine, EMBC 2009
CountryUnited States
CityMinneapolis, MN
Period09/9/209/9/6

Fingerprint

Brain-Computer Interfaces
Passive Cutaneous Anaphylaxis
Brain computer interface
Covariance matrix
Artifacts
Noise
Datasets

ASJC Scopus subject areas

  • Cell Biology
  • Developmental Biology
  • Biomedical Engineering
  • Medicine(all)

Cite this

Kawanabe, M., Vidaurre, C., Scholler, S., & Muller, K. (2009). Robust common spatial filters with a maxmin approach. In Proceedings of the 31st Annual International Conference of the IEEE Engineering in Medicine and Biology Society: Engineering the Future of Biomedicine, EMBC 2009 (pp. 2470-2473). [5334786] https://doi.org/10.1109/IEMBS.2009.5334786

Robust common spatial filters with a maxmin approach. / Kawanabe, Motoaki; Vidaurre, Carmen; Scholler, Simon; Muller, Klaus.

Proceedings of the 31st Annual International Conference of the IEEE Engineering in Medicine and Biology Society: Engineering the Future of Biomedicine, EMBC 2009. 2009. p. 2470-2473 5334786.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Kawanabe, M, Vidaurre, C, Scholler, S & Muller, K 2009, Robust common spatial filters with a maxmin approach. in Proceedings of the 31st Annual International Conference of the IEEE Engineering in Medicine and Biology Society: Engineering the Future of Biomedicine, EMBC 2009., 5334786, pp. 2470-2473, 31st Annual International Conference of the IEEE Engineering in Medicine and Biology Society: Engineering the Future of Biomedicine, EMBC 2009, Minneapolis, MN, United States, 09/9/2. https://doi.org/10.1109/IEMBS.2009.5334786
Kawanabe M, Vidaurre C, Scholler S, Muller K. Robust common spatial filters with a maxmin approach. In Proceedings of the 31st Annual International Conference of the IEEE Engineering in Medicine and Biology Society: Engineering the Future of Biomedicine, EMBC 2009. 2009. p. 2470-2473. 5334786 https://doi.org/10.1109/IEMBS.2009.5334786
Kawanabe, Motoaki ; Vidaurre, Carmen ; Scholler, Simon ; Muller, Klaus. / Robust common spatial filters with a maxmin approach. Proceedings of the 31st Annual International Conference of the IEEE Engineering in Medicine and Biology Society: Engineering the Future of Biomedicine, EMBC 2009. 2009. pp. 2470-2473
@inproceedings{1467bdfd6fdc431eb3946c1db07d3edc,
title = "Robust common spatial filters with a maxmin approach",
abstract = "Electroencephalographic signals are known to be non-stationary and easily affected by artifacts, therefore their analysis requires methods that can deal with noise. In this work we present two ways of calculating robust common spatial patterns under a maxmin approach. The worst-case objective function is optimized within prefixed sets of the covariance matrices that are defined either very simply as identity matrices or in a data driven way using PCA. We test common spatial filters derived with these two approaches with real world brain-computer interface (BCI) data sets in which we expect substantial {"}day-to-day{"} fluctuations (session transfer problem). We compare our results with the classical common spatial filters and show that both can improve the performance of the latter.",
author = "Motoaki Kawanabe and Carmen Vidaurre and Simon Scholler and Klaus Muller",
year = "2009",
month = "12",
day = "1",
doi = "10.1109/IEMBS.2009.5334786",
language = "English",
isbn = "9781424432967",
pages = "2470--2473",
booktitle = "Proceedings of the 31st Annual International Conference of the IEEE Engineering in Medicine and Biology Society: Engineering the Future of Biomedicine, EMBC 2009",

}

TY - GEN

T1 - Robust common spatial filters with a maxmin approach

AU - Kawanabe, Motoaki

AU - Vidaurre, Carmen

AU - Scholler, Simon

AU - Muller, Klaus

PY - 2009/12/1

Y1 - 2009/12/1

N2 - Electroencephalographic signals are known to be non-stationary and easily affected by artifacts, therefore their analysis requires methods that can deal with noise. In this work we present two ways of calculating robust common spatial patterns under a maxmin approach. The worst-case objective function is optimized within prefixed sets of the covariance matrices that are defined either very simply as identity matrices or in a data driven way using PCA. We test common spatial filters derived with these two approaches with real world brain-computer interface (BCI) data sets in which we expect substantial "day-to-day" fluctuations (session transfer problem). We compare our results with the classical common spatial filters and show that both can improve the performance of the latter.

AB - Electroencephalographic signals are known to be non-stationary and easily affected by artifacts, therefore their analysis requires methods that can deal with noise. In this work we present two ways of calculating robust common spatial patterns under a maxmin approach. The worst-case objective function is optimized within prefixed sets of the covariance matrices that are defined either very simply as identity matrices or in a data driven way using PCA. We test common spatial filters derived with these two approaches with real world brain-computer interface (BCI) data sets in which we expect substantial "day-to-day" fluctuations (session transfer problem). We compare our results with the classical common spatial filters and show that both can improve the performance of the latter.

UR - http://www.scopus.com/inward/record.url?scp=77950963236&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77950963236&partnerID=8YFLogxK

U2 - 10.1109/IEMBS.2009.5334786

DO - 10.1109/IEMBS.2009.5334786

M3 - Conference contribution

C2 - 19964963

AN - SCOPUS:77950963236

SN - 9781424432967

SP - 2470

EP - 2473

BT - Proceedings of the 31st Annual International Conference of the IEEE Engineering in Medicine and Biology Society: Engineering the Future of Biomedicine, EMBC 2009

ER -