Interpreting Neural Activity Through Linear and Nonlinear Models for Brain Machine Interfaces

Justin C. Sanchez, Deniz Erdogmus, Yadunandana Rao, Sung Phil Kim, Miguel Nicolelis, Johan Wessberg, Jose C. Principe

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

9 Citations (Scopus)

Abstract

Brain machine interface (BMI) design can be achieved by training linear and nonlinear models with simultaneously recorded cortical neural activity and behavior (typically the hand position of a primate). We propose the use of optimized BMI models for analyzing neural activity to assess the role of individual neurons and cortical areas in generating the performed movement. Two models (linear-feedforward and nonlinear-feedback) are trained to predict the hand position of a primate from neural recordings in a reaching task. Qualitative and quantitative investigation of the effect of neurons and their corresponding cortical areas through both models yields conclusions consistent with neurophysiologic knowledge. In addition, this analysis revealed the role of these areas and the importance of the neurons in terms of BMI design.

Original languageEnglish (US)
Title of host publicationAnnual International Conference of the IEEE Engineering in Medicine and Biology - Proceedings
EditorsR.S. Leder
Pages2160-2163
Number of pages4
Volume3
StatePublished - 2003
Externally publishedYes
EventA New Beginning for Human Health: Proceedings of the 25th Annual International Conference of the IEEE Engineering in Medicine and Biology Society - Cancun, Mexico
Duration: Sep 17 2003Sep 21 2003

Other

OtherA New Beginning for Human Health: Proceedings of the 25th Annual International Conference of the IEEE Engineering in Medicine and Biology Society
CountryMexico
CityCancun
Period9/17/039/21/03

Fingerprint

Brain
Neurons
Nonlinear feedback
Primates

Keywords

  • Analysis of neural activity
  • Brain machine interface
  • Linear/nonlinear models
  • Recurrent neural network

ASJC Scopus subject areas

  • Bioengineering

Cite this

Sanchez, J. C., Erdogmus, D., Rao, Y., Kim, S. P., Nicolelis, M., Wessberg, J., & Principe, J. C. (2003). Interpreting Neural Activity Through Linear and Nonlinear Models for Brain Machine Interfaces. In R. S. Leder (Ed.), Annual International Conference of the IEEE Engineering in Medicine and Biology - Proceedings (Vol. 3, pp. 2160-2163)

Interpreting Neural Activity Through Linear and Nonlinear Models for Brain Machine Interfaces. / Sanchez, Justin C.; Erdogmus, Deniz; Rao, Yadunandana; Kim, Sung Phil; Nicolelis, Miguel; Wessberg, Johan; Principe, Jose C.

Annual International Conference of the IEEE Engineering in Medicine and Biology - Proceedings. ed. / R.S. Leder. Vol. 3 2003. p. 2160-2163.

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

Sanchez, JC, Erdogmus, D, Rao, Y, Kim, SP, Nicolelis, M, Wessberg, J & Principe, JC 2003, Interpreting Neural Activity Through Linear and Nonlinear Models for Brain Machine Interfaces. in RS Leder (ed.), Annual International Conference of the IEEE Engineering in Medicine and Biology - Proceedings. vol. 3, pp. 2160-2163, A New Beginning for Human Health: Proceedings of the 25th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, Cancun, Mexico, 9/17/03.
Sanchez JC, Erdogmus D, Rao Y, Kim SP, Nicolelis M, Wessberg J et al. Interpreting Neural Activity Through Linear and Nonlinear Models for Brain Machine Interfaces. In Leder RS, editor, Annual International Conference of the IEEE Engineering in Medicine and Biology - Proceedings. Vol. 3. 2003. p. 2160-2163
Sanchez, Justin C. ; Erdogmus, Deniz ; Rao, Yadunandana ; Kim, Sung Phil ; Nicolelis, Miguel ; Wessberg, Johan ; Principe, Jose C. / Interpreting Neural Activity Through Linear and Nonlinear Models for Brain Machine Interfaces. Annual International Conference of the IEEE Engineering in Medicine and Biology - Proceedings. editor / R.S. Leder. Vol. 3 2003. pp. 2160-2163
@inproceedings{9eb2ed1edfb7451fa2f737c4c78c37d6,
title = "Interpreting Neural Activity Through Linear and Nonlinear Models for Brain Machine Interfaces",
abstract = "Brain machine interface (BMI) design can be achieved by training linear and nonlinear models with simultaneously recorded cortical neural activity and behavior (typically the hand position of a primate). We propose the use of optimized BMI models for analyzing neural activity to assess the role of individual neurons and cortical areas in generating the performed movement. Two models (linear-feedforward and nonlinear-feedback) are trained to predict the hand position of a primate from neural recordings in a reaching task. Qualitative and quantitative investigation of the effect of neurons and their corresponding cortical areas through both models yields conclusions consistent with neurophysiologic knowledge. In addition, this analysis revealed the role of these areas and the importance of the neurons in terms of BMI design.",
keywords = "Analysis of neural activity, Brain machine interface, Linear/nonlinear models, Recurrent neural network",
author = "Sanchez, {Justin C.} and Deniz Erdogmus and Yadunandana Rao and Kim, {Sung Phil} and Miguel Nicolelis and Johan Wessberg and Principe, {Jose C.}",
year = "2003",
language = "English (US)",
volume = "3",
pages = "2160--2163",
editor = "R.S. Leder",
booktitle = "Annual International Conference of the IEEE Engineering in Medicine and Biology - Proceedings",

}

TY - GEN

T1 - Interpreting Neural Activity Through Linear and Nonlinear Models for Brain Machine Interfaces

AU - Sanchez, Justin C.

AU - Erdogmus, Deniz

AU - Rao, Yadunandana

AU - Kim, Sung Phil

AU - Nicolelis, Miguel

AU - Wessberg, Johan

AU - Principe, Jose C.

PY - 2003

Y1 - 2003

N2 - Brain machine interface (BMI) design can be achieved by training linear and nonlinear models with simultaneously recorded cortical neural activity and behavior (typically the hand position of a primate). We propose the use of optimized BMI models for analyzing neural activity to assess the role of individual neurons and cortical areas in generating the performed movement. Two models (linear-feedforward and nonlinear-feedback) are trained to predict the hand position of a primate from neural recordings in a reaching task. Qualitative and quantitative investigation of the effect of neurons and their corresponding cortical areas through both models yields conclusions consistent with neurophysiologic knowledge. In addition, this analysis revealed the role of these areas and the importance of the neurons in terms of BMI design.

AB - Brain machine interface (BMI) design can be achieved by training linear and nonlinear models with simultaneously recorded cortical neural activity and behavior (typically the hand position of a primate). We propose the use of optimized BMI models for analyzing neural activity to assess the role of individual neurons and cortical areas in generating the performed movement. Two models (linear-feedforward and nonlinear-feedback) are trained to predict the hand position of a primate from neural recordings in a reaching task. Qualitative and quantitative investigation of the effect of neurons and their corresponding cortical areas through both models yields conclusions consistent with neurophysiologic knowledge. In addition, this analysis revealed the role of these areas and the importance of the neurons in terms of BMI design.

KW - Analysis of neural activity

KW - Brain machine interface

KW - Linear/nonlinear models

KW - Recurrent neural network

UR - http://www.scopus.com/inward/record.url?scp=1542271325&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=1542271325&partnerID=8YFLogxK

M3 - Conference contribution

VL - 3

SP - 2160

EP - 2163

BT - Annual International Conference of the IEEE Engineering in Medicine and Biology - Proceedings

A2 - Leder, R.S.

ER -