Improving BCI Performance after Classification

D. Plass - Oude Bos, Hayrettin Gürkök, B. Reuderink, Mannes Poel

  • 3 Citations

Abstract

Brain-computer interfaces offer a valuable input modality, which unfortunately comes also with a high degree of uncertainty. There are simple methods to improve detection accuracy after the incoming brain activity has already been classified, which can be divided into (1) gathering additional evidence from other sources of information, and (2) transforming the unstable classification results to be more easy to control. The methods described are easy to implement, but it is essential to apply them in the right way. This paper provides an overview of the different techniques, showing where to apply them and comparing the effects. Detection accuracy is important, but there are trade-offs to consider. Future research should investigate the effectiveness of these methods in their context of use, as well as the optimal settings to obtain the right balance between functionality and meeting the user's expectations for maximum acceptance.
Original languageUndefined
Title of host publicationProceedings of the 14th ACM international conference on Multimodal interaction, ICMI 2012
Place of PublicationNew York
PublisherACM
Pages587-594
Number of pages8
ISBN (Print)978-1-4503-1467-1
DOIs
StatePublished - Oct 2012
Event14th International Conference on Multimodal Interaction, ICMI 2012 - Santa Monica, United States

Publication series

Name
PublisherACM

Conference

Conference14th International Conference on Multimodal Interaction, ICMI 2012
Abbreviated titleICMI
CountryUnited States
CitySanta Monica
Period22/10/1226/10/12

Fingerprint

Brain computer interface
Brain

Keywords

  • HMI-MI: MULTIMODAL INTERACTIONS
  • IR-83667
  • debouncing
  • macro
  • predictive model
  • smoothing
  • EWI-22918
  • Brain-Computer Interfaces
  • Context
  • Hysteresis
  • Multimodal
  • dwell time
  • Post processing
  • METIS-296231
  • HMI-HF: Human Factors

Cite this

Plass - Oude Bos, D., Gürkök, H., Reuderink, B., & Poel, M. (2012). Improving BCI Performance after Classification. In Proceedings of the 14th ACM international conference on Multimodal interaction, ICMI 2012 (pp. 587-594). New York: ACM. DOI: 10.1145/2388676.2388799

Plass - Oude Bos, D.; Gürkök, Hayrettin; Reuderink, B.; Poel, Mannes / Improving BCI Performance after Classification.

Proceedings of the 14th ACM international conference on Multimodal interaction, ICMI 2012. New York : ACM, 2012. p. 587-594.

Research output: Scientific - peer-reviewConference contribution

@inbook{91332bcb7f4e4ef3b9a9106b6af57af6,
title = "Improving BCI Performance after Classification",
abstract = "Brain-computer interfaces offer a valuable input modality, which unfortunately comes also with a high degree of uncertainty. There are simple methods to improve detection accuracy after the incoming brain activity has already been classified, which can be divided into (1) gathering additional evidence from other sources of information, and (2) transforming the unstable classification results to be more easy to control. The methods described are easy to implement, but it is essential to apply them in the right way. This paper provides an overview of the different techniques, showing where to apply them and comparing the effects. Detection accuracy is important, but there are trade-offs to consider. Future research should investigate the effectiveness of these methods in their context of use, as well as the optimal settings to obtain the right balance between functionality and meeting the user's expectations for maximum acceptance.",
keywords = "HMI-MI: MULTIMODAL INTERACTIONS, IR-83667, debouncing, macro, predictive model, smoothing, EWI-22918, Brain-Computer Interfaces, Context, Hysteresis, Multimodal, dwell time, Post processing, METIS-296231, HMI-HF: Human Factors",
author = "{Plass - Oude Bos}, D. and Hayrettin Gürkök and B. Reuderink and Mannes Poel",
note = "10.1145/2388676.2388799",
year = "2012",
month = "10",
doi = "10.1145/2388676.2388799",
isbn = "978-1-4503-1467-1",
publisher = "ACM",
pages = "587--594",
booktitle = "Proceedings of the 14th ACM international conference on Multimodal interaction, ICMI 2012",

}

Plass - Oude Bos, D, Gürkök, H, Reuderink, B & Poel, M 2012, Improving BCI Performance after Classification. in Proceedings of the 14th ACM international conference on Multimodal interaction, ICMI 2012. ACM, New York, pp. 587-594, 14th International Conference on Multimodal Interaction, ICMI 2012, Santa Monica, United States, 22-26 October. DOI: 10.1145/2388676.2388799

Improving BCI Performance after Classification. / Plass - Oude Bos, D.; Gürkök, Hayrettin; Reuderink, B.; Poel, Mannes.

Proceedings of the 14th ACM international conference on Multimodal interaction, ICMI 2012. New York : ACM, 2012. p. 587-594.

Research output: Scientific - peer-reviewConference contribution

TY - CHAP

T1 - Improving BCI Performance after Classification

AU - Plass - Oude Bos,D.

AU - Gürkök,Hayrettin

AU - Reuderink,B.

AU - Poel,Mannes

N1 - 10.1145/2388676.2388799

PY - 2012/10

Y1 - 2012/10

N2 - Brain-computer interfaces offer a valuable input modality, which unfortunately comes also with a high degree of uncertainty. There are simple methods to improve detection accuracy after the incoming brain activity has already been classified, which can be divided into (1) gathering additional evidence from other sources of information, and (2) transforming the unstable classification results to be more easy to control. The methods described are easy to implement, but it is essential to apply them in the right way. This paper provides an overview of the different techniques, showing where to apply them and comparing the effects. Detection accuracy is important, but there are trade-offs to consider. Future research should investigate the effectiveness of these methods in their context of use, as well as the optimal settings to obtain the right balance between functionality and meeting the user's expectations for maximum acceptance.

AB - Brain-computer interfaces offer a valuable input modality, which unfortunately comes also with a high degree of uncertainty. There are simple methods to improve detection accuracy after the incoming brain activity has already been classified, which can be divided into (1) gathering additional evidence from other sources of information, and (2) transforming the unstable classification results to be more easy to control. The methods described are easy to implement, but it is essential to apply them in the right way. This paper provides an overview of the different techniques, showing where to apply them and comparing the effects. Detection accuracy is important, but there are trade-offs to consider. Future research should investigate the effectiveness of these methods in their context of use, as well as the optimal settings to obtain the right balance between functionality and meeting the user's expectations for maximum acceptance.

KW - HMI-MI: MULTIMODAL INTERACTIONS

KW - IR-83667

KW - debouncing

KW - macro

KW - predictive model

KW - smoothing

KW - EWI-22918

KW - Brain-Computer Interfaces

KW - Context

KW - Hysteresis

KW - Multimodal

KW - dwell time

KW - Post processing

KW - METIS-296231

KW - HMI-HF: Human Factors

U2 - 10.1145/2388676.2388799

DO - 10.1145/2388676.2388799

M3 - Conference contribution

SN - 978-1-4503-1467-1

SP - 587

EP - 594

BT - Proceedings of the 14th ACM international conference on Multimodal interaction, ICMI 2012

PB - ACM

ER -

Plass - Oude Bos D, Gürkök H, Reuderink B, Poel M. Improving BCI Performance after Classification. In Proceedings of the 14th ACM international conference on Multimodal interaction, ICMI 2012. New York: ACM. 2012. p. 587-594. Available from, DOI: 10.1145/2388676.2388799