Adaptive CSP for user independence in MI-BCI paradigm for upper limb stroke rehabilitation

Ana P. Costa, Jakob Skadkær Møller, Helle K. Iversen, Sadasivan Puthusserypady

Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review

351 Downloads (Pure)

Abstract

A 3-class motor imagery (MI) Brain-Computer Interface (BCI) system, that implements subject adaptation with short to non-existing calibration sessions is proposed. The proposed adaptive common spatial patterns (ACSP) algorithm was tested on two datasets (an open source data set (4-class MI), and an in-house data set (3-class MI)). Results show that when longcalibrationdataisavailable,theACSPperformsonlyslightly better (4%) than the CSP, but for short calibration sessions, the ACSP significantly improved the performance (up to 4-fold). An investigation into class separability of the in-house data set was performed and was concluded that the “Pinch”movement was more easily discriminated than “Grasp” and “Elbow Flexion”. The proposed paradigm proved feasible and provided insights to help choose the motor tasks leading to best results in potential real-life applications. The ACSP enabled a successful semi user independent scenario and showed potential to be a tool towards an improved, personalized stroke rehabilitation protocol.
Original languageEnglish
Title of host publicationProceedings of GlobalSIP 2018
PublisherIEEE
Publication date2019
Pages420-423
ISBN (Print)978-1-7281-1295-4
Publication statusPublished - 2019
Event6th IEEE Global Conference on Signal and Information Processing - The Disneyland Hotel, Anaheim, United States
Duration: 26 Nov 201829 Nov 2018
Conference number: 6
https://2018.ieeeglobalsip.org/default.asp

Conference

Conference6th IEEE Global Conference on Signal and Information Processing
Number6
LocationThe Disneyland Hotel
CountryUnited States
CityAnaheim
Period26/11/201829/11/2018
Internet address

Keywords

  • Brain-computer interface (BCI)
  • Stroke rehabilitation
  • Sensorimotor rhythms (SMR)
  • Adaptive Common Spatial Patterns (ACSP)

Cite this

Costa, A. P., Møller, J. S., Iversen, H. K., & Puthusserypady, S. (2019). Adaptive CSP for user independence in MI-BCI paradigm for upper limb stroke rehabilitation. In Proceedings of GlobalSIP 2018 (pp. 420-423). IEEE.
Costa, Ana P. ; Møller, Jakob Skadkær ; Iversen, Helle K. ; Puthusserypady, Sadasivan. / Adaptive CSP for user independence in MI-BCI paradigm for upper limb stroke rehabilitation. Proceedings of GlobalSIP 2018. IEEE, 2019. pp. 420-423
@inproceedings{cfd82eb04f04490886e089fa5967b774,
title = "Adaptive CSP for user independence in MI-BCI paradigm for upper limb stroke rehabilitation",
abstract = "A 3-class motor imagery (MI) Brain-Computer Interface (BCI) system, that implements subject adaptation with short to non-existing calibration sessions is proposed. The proposed adaptive common spatial patterns (ACSP) algorithm was tested on two datasets (an open source data set (4-class MI), and an in-house data set (3-class MI)). Results show that when longcalibrationdataisavailable,theACSPperformsonlyslightly better (4{\%}) than the CSP, but for short calibration sessions, the ACSP significantly improved the performance (up to 4-fold). An investigation into class separability of the in-house data set was performed and was concluded that the “Pinch”movement was more easily discriminated than “Grasp” and “Elbow Flexion”. The proposed paradigm proved feasible and provided insights to help choose the motor tasks leading to best results in potential real-life applications. The ACSP enabled a successful semi user independent scenario and showed potential to be a tool towards an improved, personalized stroke rehabilitation protocol.",
keywords = "Brain-computer interface (BCI), Stroke rehabilitation, Sensorimotor rhythms (SMR), Adaptive Common Spatial Patterns (ACSP)",
author = "Costa, {Ana P.} and M{\o}ller, {Jakob Skadk{\ae}r} and Iversen, {Helle K.} and Sadasivan Puthusserypady",
year = "2019",
language = "English",
isbn = "978-1-7281-1295-4",
pages = "420--423",
booktitle = "Proceedings of GlobalSIP 2018",
publisher = "IEEE",
address = "United States",

}

Costa, AP, Møller, JS, Iversen, HK & Puthusserypady, S 2019, Adaptive CSP for user independence in MI-BCI paradigm for upper limb stroke rehabilitation. in Proceedings of GlobalSIP 2018. IEEE, pp. 420-423, 6th IEEE Global Conference on Signal and Information Processing, Anaheim, United States, 26/11/2018.

Adaptive CSP for user independence in MI-BCI paradigm for upper limb stroke rehabilitation. / Costa, Ana P.; Møller, Jakob Skadkær; Iversen, Helle K.; Puthusserypady, Sadasivan.

Proceedings of GlobalSIP 2018. IEEE, 2019. p. 420-423.

Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review

TY - GEN

T1 - Adaptive CSP for user independence in MI-BCI paradigm for upper limb stroke rehabilitation

AU - Costa, Ana P.

AU - Møller, Jakob Skadkær

AU - Iversen, Helle K.

AU - Puthusserypady, Sadasivan

PY - 2019

Y1 - 2019

N2 - A 3-class motor imagery (MI) Brain-Computer Interface (BCI) system, that implements subject adaptation with short to non-existing calibration sessions is proposed. The proposed adaptive common spatial patterns (ACSP) algorithm was tested on two datasets (an open source data set (4-class MI), and an in-house data set (3-class MI)). Results show that when longcalibrationdataisavailable,theACSPperformsonlyslightly better (4%) than the CSP, but for short calibration sessions, the ACSP significantly improved the performance (up to 4-fold). An investigation into class separability of the in-house data set was performed and was concluded that the “Pinch”movement was more easily discriminated than “Grasp” and “Elbow Flexion”. The proposed paradigm proved feasible and provided insights to help choose the motor tasks leading to best results in potential real-life applications. The ACSP enabled a successful semi user independent scenario and showed potential to be a tool towards an improved, personalized stroke rehabilitation protocol.

AB - A 3-class motor imagery (MI) Brain-Computer Interface (BCI) system, that implements subject adaptation with short to non-existing calibration sessions is proposed. The proposed adaptive common spatial patterns (ACSP) algorithm was tested on two datasets (an open source data set (4-class MI), and an in-house data set (3-class MI)). Results show that when longcalibrationdataisavailable,theACSPperformsonlyslightly better (4%) than the CSP, but for short calibration sessions, the ACSP significantly improved the performance (up to 4-fold). An investigation into class separability of the in-house data set was performed and was concluded that the “Pinch”movement was more easily discriminated than “Grasp” and “Elbow Flexion”. The proposed paradigm proved feasible and provided insights to help choose the motor tasks leading to best results in potential real-life applications. The ACSP enabled a successful semi user independent scenario and showed potential to be a tool towards an improved, personalized stroke rehabilitation protocol.

KW - Brain-computer interface (BCI)

KW - Stroke rehabilitation

KW - Sensorimotor rhythms (SMR)

KW - Adaptive Common Spatial Patterns (ACSP)

M3 - Article in proceedings

SN - 978-1-7281-1295-4

SP - 420

EP - 423

BT - Proceedings of GlobalSIP 2018

PB - IEEE

ER -

Costa AP, Møller JS, Iversen HK, Puthusserypady S. Adaptive CSP for user independence in MI-BCI paradigm for upper limb stroke rehabilitation. In Proceedings of GlobalSIP 2018. IEEE. 2019. p. 420-423