Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Tensor decomposition of EEG signals for transfer learning applications

Karlsson, Linda LU orcid ; Fallenius, Emma ; Bergeling, Carolina LU orcid and Bernhardsson, Bo LU orcid (2024) In Brain-Computer Interfaces 11(4). p.178-192
Abstract

We address the recognized person-to-person Brain–Computer Interface (BCI) calibration problem and tackle session-dependency through the use of unsupervised canonical polyadic (CP) tensor decomposition. For a motor imagery task, the approach reveals universal structures within EEG data, common between subjects and prominent for a certain task. Further, we develop a novel similarity measure that includes weighting of the decomposition’s factor matrices, and argue that it is more representative than what has previously been presented in literature. The proposed similarity measure shows potential in a BCI classification task, i.e. drowsiness during simulated driving (average Pearson correlation of 0.6).

Please use this url to cite or link to this publication:
author
; ; and
organization
publishing date
type
Contribution to journal
publication status
published
subject
keywords
BCI calibration, Brain–computer interface, similarity measure, tensor decomposition, transfer learning
in
Brain-Computer Interfaces
volume
11
issue
4
pages
178 - 192
publisher
Taylor & Francis
external identifiers
  • scopus:85205544184
ISSN
2326-263X
DOI
10.1080/2326263X.2024.2401663
project
Optimizing the Next Generation Brain Computer Interfaces using Cloud Computing
language
English
LU publication?
yes
additional info
Publisher Copyright: © 2024 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group.
id
c706e5c9-e100-40ad-96dd-dd55be01abc3
date added to LUP
2024-10-15 11:41:09
date last changed
2025-06-25 08:22:32
@article{c706e5c9-e100-40ad-96dd-dd55be01abc3,
  abstract     = {{<p>We address the recognized person-to-person Brain–Computer Interface (BCI) calibration problem and tackle session-dependency through the use of unsupervised canonical polyadic (CP) tensor decomposition. For a motor imagery task, the approach reveals universal structures within EEG data, common between subjects and prominent for a certain task. Further, we develop a novel similarity measure that includes weighting of the decomposition’s factor matrices, and argue that it is more representative than what has previously been presented in literature. The proposed similarity measure shows potential in a BCI classification task, i.e. drowsiness during simulated driving (average Pearson correlation of 0.6).</p>}},
  author       = {{Karlsson, Linda and Fallenius, Emma and Bergeling, Carolina and Bernhardsson, Bo}},
  issn         = {{2326-263X}},
  keywords     = {{BCI calibration; Brain–computer interface; similarity measure; tensor decomposition; transfer learning}},
  language     = {{eng}},
  number       = {{4}},
  pages        = {{178--192}},
  publisher    = {{Taylor & Francis}},
  series       = {{Brain-Computer Interfaces}},
  title        = {{Tensor decomposition of EEG signals for transfer learning applications}},
  url          = {{http://dx.doi.org/10.1080/2326263X.2024.2401663}},
  doi          = {{10.1080/2326263X.2024.2401663}},
  volume       = {{11}},
  year         = {{2024}},
}