A learning scheme for reach to grasp movements: On emg-based interfaces using task specific motion decoding models

Minas V. Liarokapis, Panagiotis Artemiadis, Kostas J. Kyriakopoulos, Elias S. Manolakos

Research output: Contribution to journalArticle

25 Citations (Scopus)

Abstract

A learning scheme based on random forests is used to discriminate between different reach to grasp movements in 3-D space, based on the myoelectric activity of human muscles of the upper-arm and the forearm. Task specificity for motion decoding is introduced in two different levels: Subspace to move toward and object to be grasped. The discrimination between the different reach to grasp strategies is accomplished with machine learning techniques for classification. The classification decision is then used in order to trigger an EMG-based task-specific motion decoding model. Task specific models manage to outperform 'general' models providing better estimation accuracy. Thus, the proposed scheme takes advantage of a framework incorporating both a classifier and a regressor that cooperate advantageously in order to split the task space. The proposed learning scheme can be easily used to a series of EMG-based interfaces that must operate in real time, providing data-driven capabilities for multiclass problems, that occur in everyday life complex environments.

Original languageEnglish (US)
Article number6507235
Pages (from-to)915-921
Number of pages7
JournalIEEE Journal of Biomedical and Health Informatics
Volume17
Issue number5
DOIs
StatePublished - 2013

Fingerprint

Decoding
Learning
Forearm
Human Activities
Arm
Muscles
Muscle
Learning systems
Classifiers
Discrimination (Psychology)
Machine Learning
Forests

Keywords

  • Electromyography (EMG)
  • learning scheme
  • random forests
  • task specificity

ASJC Scopus subject areas

  • Biotechnology
  • Computer Science Applications
  • Electrical and Electronic Engineering
  • Health Information Management
  • Medicine(all)

Cite this

A learning scheme for reach to grasp movements : On emg-based interfaces using task specific motion decoding models. / Liarokapis, Minas V.; Artemiadis, Panagiotis; Kyriakopoulos, Kostas J.; Manolakos, Elias S.

In: IEEE Journal of Biomedical and Health Informatics, Vol. 17, No. 5, 6507235, 2013, p. 915-921.

Research output: Contribution to journalArticle

Liarokapis, Minas V. ; Artemiadis, Panagiotis ; Kyriakopoulos, Kostas J. ; Manolakos, Elias S. / A learning scheme for reach to grasp movements : On emg-based interfaces using task specific motion decoding models. In: IEEE Journal of Biomedical and Health Informatics. 2013 ; Vol. 17, No. 5. pp. 915-921.
@article{f533124a8d92495eaac84268f93b9146,
title = "A learning scheme for reach to grasp movements: On emg-based interfaces using task specific motion decoding models",
abstract = "A learning scheme based on random forests is used to discriminate between different reach to grasp movements in 3-D space, based on the myoelectric activity of human muscles of the upper-arm and the forearm. Task specificity for motion decoding is introduced in two different levels: Subspace to move toward and object to be grasped. The discrimination between the different reach to grasp strategies is accomplished with machine learning techniques for classification. The classification decision is then used in order to trigger an EMG-based task-specific motion decoding model. Task specific models manage to outperform 'general' models providing better estimation accuracy. Thus, the proposed scheme takes advantage of a framework incorporating both a classifier and a regressor that cooperate advantageously in order to split the task space. The proposed learning scheme can be easily used to a series of EMG-based interfaces that must operate in real time, providing data-driven capabilities for multiclass problems, that occur in everyday life complex environments.",
keywords = "Electromyography (EMG), learning scheme, random forests, task specificity",
author = "Liarokapis, {Minas V.} and Panagiotis Artemiadis and Kyriakopoulos, {Kostas J.} and Manolakos, {Elias S.}",
year = "2013",
doi = "10.1109/JBHI.2013.2259594",
language = "English (US)",
volume = "17",
pages = "915--921",
journal = "IEEE Journal of Biomedical and Health Informatics",
issn = "2168-2194",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "5",

}

TY - JOUR

T1 - A learning scheme for reach to grasp movements

T2 - On emg-based interfaces using task specific motion decoding models

AU - Liarokapis, Minas V.

AU - Artemiadis, Panagiotis

AU - Kyriakopoulos, Kostas J.

AU - Manolakos, Elias S.

PY - 2013

Y1 - 2013

N2 - A learning scheme based on random forests is used to discriminate between different reach to grasp movements in 3-D space, based on the myoelectric activity of human muscles of the upper-arm and the forearm. Task specificity for motion decoding is introduced in two different levels: Subspace to move toward and object to be grasped. The discrimination between the different reach to grasp strategies is accomplished with machine learning techniques for classification. The classification decision is then used in order to trigger an EMG-based task-specific motion decoding model. Task specific models manage to outperform 'general' models providing better estimation accuracy. Thus, the proposed scheme takes advantage of a framework incorporating both a classifier and a regressor that cooperate advantageously in order to split the task space. The proposed learning scheme can be easily used to a series of EMG-based interfaces that must operate in real time, providing data-driven capabilities for multiclass problems, that occur in everyday life complex environments.

AB - A learning scheme based on random forests is used to discriminate between different reach to grasp movements in 3-D space, based on the myoelectric activity of human muscles of the upper-arm and the forearm. Task specificity for motion decoding is introduced in two different levels: Subspace to move toward and object to be grasped. The discrimination between the different reach to grasp strategies is accomplished with machine learning techniques for classification. The classification decision is then used in order to trigger an EMG-based task-specific motion decoding model. Task specific models manage to outperform 'general' models providing better estimation accuracy. Thus, the proposed scheme takes advantage of a framework incorporating both a classifier and a regressor that cooperate advantageously in order to split the task space. The proposed learning scheme can be easily used to a series of EMG-based interfaces that must operate in real time, providing data-driven capabilities for multiclass problems, that occur in everyday life complex environments.

KW - Electromyography (EMG)

KW - learning scheme

KW - random forests

KW - task specificity

UR - http://www.scopus.com/inward/record.url?scp=84883887265&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84883887265&partnerID=8YFLogxK

U2 - 10.1109/JBHI.2013.2259594

DO - 10.1109/JBHI.2013.2259594

M3 - Article

C2 - 25055370

AN - SCOPUS:84883887265

VL - 17

SP - 915

EP - 921

JO - IEEE Journal of Biomedical and Health Informatics

JF - IEEE Journal of Biomedical and Health Informatics

SN - 2168-2194

IS - 5

M1 - 6507235

ER -