Current source density estimates improve the discriminability of scalp-level brain connectivity features related to motor-imagery tasks

Dheeraj Rathee, Hubert Cecotti, Girijesh Prasad

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

Abstract

Recent progress in the number of studies involving
brain connectivity analysis of motor imagery (MI) tasks for
brain-computer interface (BCI) systems has warranted the
need for pre-processing methods. The objective of this study
is to evaluate the impact of current source density (CSD)
estimation from raw electroencephalogram (EEG) signals on
the classification performance of scalp level brain connectivity feature based MI-BCI. In particular, time-domain partial
Granger causality (PGC) method was implemented on the raw
EEG signals and CSD signals of a publicly available dataset
for the estimation of brain connectivity features. Moreover,
pairwise binary classifications of four different MI tasks were
performed in inter-session and intra-session conditions using a
support vector machine classifier. The results showed that CSD
provided a statistically significant increase of the AUC: 20.28%
in the inter-session condition; 12.54% and 13.92% with session
01 and session 02, respectively, in the intra-session condition.
These results show that pre-processing of EEG signals is crucial
for single-trial connectivity features based MI-BCI systems and
CSD can enhance their overall performance.
LanguageEnglish
Title of host publicationProc. 2018 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC)
Number of pages4
ISBN (Electronic)978-1-5386-3646-6
Publication statusPublished - 29 Oct 2018
Event2018 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC) - Honolulu, HI, United States
Duration: 18 Jul 201921 Jul 2019
http://10.1109/EMBC.2018.8513417

Conference

Conference2018 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC)
CountryUnited States
CityHonolulu, HI
Period18/07/1921/07/19
Internet address

Fingerprint

Brain
Electroencephalography
Processing
Interfaces (computer)
Classifiers

Keywords

  • Current source density
  • scalp-level brain connectivity features
  • motor-imagery

Cite this

Rathee, D., Cecotti, H., & Prasad, G. (2018). Current source density estimates improve the discriminability of scalp-level brain connectivity features related to motor-imagery tasks. In Proc. 2018 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC)
Rathee, Dheeraj ; Cecotti, Hubert ; Prasad, Girijesh. / Current source density estimates improve the discriminability of scalp-level brain connectivity features related to motor-imagery tasks. Proc. 2018 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC). 2018.
@inproceedings{a59d7aa4e8f448d890a6f49760d3566c,
title = "Current source density estimates improve the discriminability of scalp-level brain connectivity features related to motor-imagery tasks",
abstract = "Recent progress in the number of studies involvingbrain connectivity analysis of motor imagery (MI) tasks forbrain-computer interface (BCI) systems has warranted theneed for pre-processing methods. The objective of this studyis to evaluate the impact of current source density (CSD)estimation from raw electroencephalogram (EEG) signals onthe classification performance of scalp level brain connectivity feature based MI-BCI. In particular, time-domain partialGranger causality (PGC) method was implemented on the rawEEG signals and CSD signals of a publicly available datasetfor the estimation of brain connectivity features. Moreover,pairwise binary classifications of four different MI tasks wereperformed in inter-session and intra-session conditions using asupport vector machine classifier. The results showed that CSDprovided a statistically significant increase of the AUC: 20.28{\%}in the inter-session condition; 12.54{\%} and 13.92{\%} with session01 and session 02, respectively, in the intra-session condition.These results show that pre-processing of EEG signals is crucialfor single-trial connectivity features based MI-BCI systems andCSD can enhance their overall performance.",
keywords = "Current source density, scalp-level brain connectivity features, motor-imagery",
author = "Dheeraj Rathee and Hubert Cecotti and Girijesh Prasad",
year = "2018",
month = "10",
day = "29",
language = "English",
isbn = "978-1-5386-3647-3",
booktitle = "Proc. 2018 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC)",

}

Rathee, D, Cecotti, H & Prasad, G 2018, Current source density estimates improve the discriminability of scalp-level brain connectivity features related to motor-imagery tasks. in Proc. 2018 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC). 2018 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), Honolulu, HI, United States, 18/07/19.

Current source density estimates improve the discriminability of scalp-level brain connectivity features related to motor-imagery tasks. / Rathee, Dheeraj; Cecotti, Hubert; Prasad, Girijesh.

Proc. 2018 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC). 2018.

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

TY - GEN

T1 - Current source density estimates improve the discriminability of scalp-level brain connectivity features related to motor-imagery tasks

AU - Rathee, Dheeraj

AU - Cecotti, Hubert

AU - Prasad, Girijesh

PY - 2018/10/29

Y1 - 2018/10/29

N2 - Recent progress in the number of studies involvingbrain connectivity analysis of motor imagery (MI) tasks forbrain-computer interface (BCI) systems has warranted theneed for pre-processing methods. The objective of this studyis to evaluate the impact of current source density (CSD)estimation from raw electroencephalogram (EEG) signals onthe classification performance of scalp level brain connectivity feature based MI-BCI. In particular, time-domain partialGranger causality (PGC) method was implemented on the rawEEG signals and CSD signals of a publicly available datasetfor the estimation of brain connectivity features. Moreover,pairwise binary classifications of four different MI tasks wereperformed in inter-session and intra-session conditions using asupport vector machine classifier. The results showed that CSDprovided a statistically significant increase of the AUC: 20.28%in the inter-session condition; 12.54% and 13.92% with session01 and session 02, respectively, in the intra-session condition.These results show that pre-processing of EEG signals is crucialfor single-trial connectivity features based MI-BCI systems andCSD can enhance their overall performance.

AB - Recent progress in the number of studies involvingbrain connectivity analysis of motor imagery (MI) tasks forbrain-computer interface (BCI) systems has warranted theneed for pre-processing methods. The objective of this studyis to evaluate the impact of current source density (CSD)estimation from raw electroencephalogram (EEG) signals onthe classification performance of scalp level brain connectivity feature based MI-BCI. In particular, time-domain partialGranger causality (PGC) method was implemented on the rawEEG signals and CSD signals of a publicly available datasetfor the estimation of brain connectivity features. Moreover,pairwise binary classifications of four different MI tasks wereperformed in inter-session and intra-session conditions using asupport vector machine classifier. The results showed that CSDprovided a statistically significant increase of the AUC: 20.28%in the inter-session condition; 12.54% and 13.92% with session01 and session 02, respectively, in the intra-session condition.These results show that pre-processing of EEG signals is crucialfor single-trial connectivity features based MI-BCI systems andCSD can enhance their overall performance.

KW - Current source density

KW - scalp-level brain connectivity features

KW - motor-imagery

UR - http://10.1109/EMBC.2018.8513417

M3 - Conference contribution

SN - 978-1-5386-3647-3

BT - Proc. 2018 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC)

ER -

Rathee D, Cecotti H, Prasad G. Current source density estimates improve the discriminability of scalp-level brain connectivity features related to motor-imagery tasks. In Proc. 2018 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC). 2018