Mapping broadband electrocorticographic recordings to two-dimensional hand trajectories in humans. Motor control features

Aysegul Gunduz, Justin C. Sanchez, Paul R. Carney, Jose C. Principe

Research output: Contribution to journalArticle

25 Citations (Scopus)

Abstract

Brain-machine interfaces (BMIs) aim to translate the motor intent of locked-in patients into neuroprosthetic control commands. Electrocorticographical (ECoG) signals provide promising neural inputs to BMIs as shown in recent studies. In this paper, we utilize a broadband spectrum above the fast gamma ranges and systematically study the role of spectral resolution, in which the broadband is partitioned, on the reconstruction of the patients' hand trajectories. Traditionally, the power of ECoG rhythms (<200-300 Hz) has been computed in short duration bins and instantaneously and linearly mapped to cursor trajectories. Neither time embedding, nor nonlinear mappings have been previously implemented in ECoG neuroprosthesis. Herein, mapping of neural modulations to goal-oriented motor behavior is achieved via linear adaptive filters with embedded memory depths and as a novelty through echo state networks (ESNs), which provide nonlinear mappings without compromising training complexity or increasing the number of model parameters, with up to 85% correlation. Reconstructed hand trajectories are analyzed through spatial, spectral and temporal sensitivities. The superiority of nonlinear mappings in the cases of low spectral resolution and abundance of interictal activity is discussed.

Original languageEnglish
Pages (from-to)1257-1270
Number of pages14
JournalNeural Networks
Volume22
Issue number9
DOIs
StatePublished - Nov 1 2009
Externally publishedYes

Fingerprint

Brain-Computer Interfaces
Hand
Trajectories
Spectral resolution
Brain
Adaptive filters
Bins
Modulation
Data storage equipment
Power (Psychology)

Keywords

  • Brain-machine interfaces
  • Electrocorticography
  • Human neuroprosthesis
  • Motor control
  • Neural decoding

ASJC Scopus subject areas

  • Artificial Intelligence
  • Cognitive Neuroscience

Cite this

Mapping broadband electrocorticographic recordings to two-dimensional hand trajectories in humans. Motor control features. / Gunduz, Aysegul; Sanchez, Justin C.; Carney, Paul R.; Principe, Jose C.

In: Neural Networks, Vol. 22, No. 9, 01.11.2009, p. 1257-1270.

Research output: Contribution to journalArticle

Gunduz, Aysegul ; Sanchez, Justin C. ; Carney, Paul R. ; Principe, Jose C. / Mapping broadband electrocorticographic recordings to two-dimensional hand trajectories in humans. Motor control features. In: Neural Networks. 2009 ; Vol. 22, No. 9. pp. 1257-1270.
@article{dc4f0c26ef8b49d09210c1666dbe8b6b,
title = "Mapping broadband electrocorticographic recordings to two-dimensional hand trajectories in humans. Motor control features",
abstract = "Brain-machine interfaces (BMIs) aim to translate the motor intent of locked-in patients into neuroprosthetic control commands. Electrocorticographical (ECoG) signals provide promising neural inputs to BMIs as shown in recent studies. In this paper, we utilize a broadband spectrum above the fast gamma ranges and systematically study the role of spectral resolution, in which the broadband is partitioned, on the reconstruction of the patients' hand trajectories. Traditionally, the power of ECoG rhythms (<200-300 Hz) has been computed in short duration bins and instantaneously and linearly mapped to cursor trajectories. Neither time embedding, nor nonlinear mappings have been previously implemented in ECoG neuroprosthesis. Herein, mapping of neural modulations to goal-oriented motor behavior is achieved via linear adaptive filters with embedded memory depths and as a novelty through echo state networks (ESNs), which provide nonlinear mappings without compromising training complexity or increasing the number of model parameters, with up to 85{\%} correlation. Reconstructed hand trajectories are analyzed through spatial, spectral and temporal sensitivities. The superiority of nonlinear mappings in the cases of low spectral resolution and abundance of interictal activity is discussed.",
keywords = "Brain-machine interfaces, Electrocorticography, Human neuroprosthesis, Motor control, Neural decoding",
author = "Aysegul Gunduz and Sanchez, {Justin C.} and Carney, {Paul R.} and Principe, {Jose C.}",
year = "2009",
month = "11",
day = "1",
doi = "10.1016/j.neunet.2009.06.036",
language = "English",
volume = "22",
pages = "1257--1270",
journal = "Neural Networks",
issn = "0893-6080",
publisher = "Elsevier Limited",
number = "9",

}

TY - JOUR

T1 - Mapping broadband electrocorticographic recordings to two-dimensional hand trajectories in humans. Motor control features

AU - Gunduz, Aysegul

AU - Sanchez, Justin C.

AU - Carney, Paul R.

AU - Principe, Jose C.

PY - 2009/11/1

Y1 - 2009/11/1

N2 - Brain-machine interfaces (BMIs) aim to translate the motor intent of locked-in patients into neuroprosthetic control commands. Electrocorticographical (ECoG) signals provide promising neural inputs to BMIs as shown in recent studies. In this paper, we utilize a broadband spectrum above the fast gamma ranges and systematically study the role of spectral resolution, in which the broadband is partitioned, on the reconstruction of the patients' hand trajectories. Traditionally, the power of ECoG rhythms (<200-300 Hz) has been computed in short duration bins and instantaneously and linearly mapped to cursor trajectories. Neither time embedding, nor nonlinear mappings have been previously implemented in ECoG neuroprosthesis. Herein, mapping of neural modulations to goal-oriented motor behavior is achieved via linear adaptive filters with embedded memory depths and as a novelty through echo state networks (ESNs), which provide nonlinear mappings without compromising training complexity or increasing the number of model parameters, with up to 85% correlation. Reconstructed hand trajectories are analyzed through spatial, spectral and temporal sensitivities. The superiority of nonlinear mappings in the cases of low spectral resolution and abundance of interictal activity is discussed.

AB - Brain-machine interfaces (BMIs) aim to translate the motor intent of locked-in patients into neuroprosthetic control commands. Electrocorticographical (ECoG) signals provide promising neural inputs to BMIs as shown in recent studies. In this paper, we utilize a broadband spectrum above the fast gamma ranges and systematically study the role of spectral resolution, in which the broadband is partitioned, on the reconstruction of the patients' hand trajectories. Traditionally, the power of ECoG rhythms (<200-300 Hz) has been computed in short duration bins and instantaneously and linearly mapped to cursor trajectories. Neither time embedding, nor nonlinear mappings have been previously implemented in ECoG neuroprosthesis. Herein, mapping of neural modulations to goal-oriented motor behavior is achieved via linear adaptive filters with embedded memory depths and as a novelty through echo state networks (ESNs), which provide nonlinear mappings without compromising training complexity or increasing the number of model parameters, with up to 85% correlation. Reconstructed hand trajectories are analyzed through spatial, spectral and temporal sensitivities. The superiority of nonlinear mappings in the cases of low spectral resolution and abundance of interictal activity is discussed.

KW - Brain-machine interfaces

KW - Electrocorticography

KW - Human neuroprosthesis

KW - Motor control

KW - Neural decoding

UR - http://www.scopus.com/inward/record.url?scp=70350228915&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=70350228915&partnerID=8YFLogxK

U2 - 10.1016/j.neunet.2009.06.036

DO - 10.1016/j.neunet.2009.06.036

M3 - Article

C2 - 19647981

AN - SCOPUS:70350228915

VL - 22

SP - 1257

EP - 1270

JO - Neural Networks

JF - Neural Networks

SN - 0893-6080

IS - 9

ER -