Reach-to-grasp motions: Towards a dynamic classification approach for upper-limp prosthesis

I. Batzianoulis; A. M. Simon; L. Hargrove; A. Billard

doi:10.1109/NER.2019.8717110

Reach-to-grasp motions: Towards a dynamic classification approach for upper-limp prosthesis

I. Batzianoulis, A. M. Simon, L. Hargrove, A. Billard

Physical Medicine and Rehabilitation

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

4 Scopus citations

Abstract

During reach-to-grasp motions, the Electromyo-graphic (EMG) activity of the arm varies depending on motion stage. The variability of the EMG signals results in low classification accuracy during the reaching phase, delaying the activation of the prosthesis. To increase the efficiency of the pattern-recognition system, we investigate the muscle activity of four individuals with below-elbow amputation performing reach-to-grasp motions and segment the arm-motion into three phases with respect to the extension of the arm. Furthermore, we model the dynamic muscle contractions of each class with Gaussian distributions over the different phases and the overall motion. We quantify of the overlap among the classes with the Hellinger distance and notice larger values and, thus, smaller overlaps among the classes with the segmentation to motion phases. A Linear Discriminant Analysis classifier with phase segmentation affects positively the classification accuracy by 6 - 10% on average.

Original language	English (US)
Title of host publication	9th International IEEE EMBS Conference on Neural Engineering, NER 2019
Publisher	IEEE Computer Society
Pages	287-290
Number of pages	4
ISBN (Electronic)	9781538679210
DOIs	https://doi.org/10.1109/NER.2019.8717110
State	Published - May 16 2019
Event	9th International IEEE EMBS Conference on Neural Engineering, NER 2019 - San Francisco, United States Duration: Mar 20 2019 → Mar 23 2019

Publication series

Name	International IEEE/EMBS Conference on Neural Engineering, NER
Volume	2019-March
ISSN (Print)	1948-3546
ISSN (Electronic)	1948-3554

Conference

Conference	9th International IEEE EMBS Conference on Neural Engineering, NER 2019
Country/Territory	United States
City	San Francisco
Period	3/20/19 → 3/23/19

Funding

*This work has received funding from the Swiss National Science Foundation through the National Centre of Competence in Research (NCCR) in Robotics, the Hasler foundation and the United States National Institute of Health 1I. Batzianoulis and A. Billard are with Learning Algorithms and Systems Laboratory (LASA) at Ecole Polytechnique Federale de Lausanne (EPFL), Switzerland 2 A. M. Simon and L. Hargrove are with the Center of Bionics Medicine at Shirley Ryan Abilitylab and the department of Mechanical Engineering of the Northwestern University, Chicago, Il, USA

ASJC Scopus subject areas

Artificial Intelligence
Mechanical Engineering

Access to Document

10.1109/NER.2019.8717110

Cite this

Batzianoulis, I., Simon, A. M., Hargrove, L., & Billard, A. (2019). Reach-to-grasp motions: Towards a dynamic classification approach for upper-limp prosthesis. In 9th International IEEE EMBS Conference on Neural Engineering, NER 2019 (pp. 287-290). Article 8717110 (International IEEE/EMBS Conference on Neural Engineering, NER; Vol. 2019-March). IEEE Computer Society. https://doi.org/10.1109/NER.2019.8717110

@inproceedings{4ca22af5fece4196a26bc8e732adece2,

title = "Reach-to-grasp motions: Towards a dynamic classification approach for upper-limp prosthesis",

abstract = "During reach-to-grasp motions, the Electromyo-graphic (EMG) activity of the arm varies depending on motion stage. The variability of the EMG signals results in low classification accuracy during the reaching phase, delaying the activation of the prosthesis. To increase the efficiency of the pattern-recognition system, we investigate the muscle activity of four individuals with below-elbow amputation performing reach-to-grasp motions and segment the arm-motion into three phases with respect to the extension of the arm. Furthermore, we model the dynamic muscle contractions of each class with Gaussian distributions over the different phases and the overall motion. We quantify of the overlap among the classes with the Hellinger distance and notice larger values and, thus, smaller overlaps among the classes with the segmentation to motion phases. A Linear Discriminant Analysis classifier with phase segmentation affects positively the classification accuracy by 6 - 10% on average.",

author = "I. Batzianoulis and Simon, {A. M.} and L. Hargrove and A. Billard",

note = "Publisher Copyright: {\textcopyright} 2019 IEEE.; 9th International IEEE EMBS Conference on Neural Engineering, NER 2019 ; Conference date: 20-03-2019 Through 23-03-2019",

year = "2019",

month = may,

day = "16",

doi = "10.1109/NER.2019.8717110",

language = "English (US)",

series = "International IEEE/EMBS Conference on Neural Engineering, NER",

publisher = "IEEE Computer Society",

pages = "287--290",

booktitle = "9th International IEEE EMBS Conference on Neural Engineering, NER 2019",

address = "United States",

}

Batzianoulis, I, Simon, AM , Hargrove, L & Billard, A 2019, Reach-to-grasp motions: Towards a dynamic classification approach for upper-limp prosthesis. in 9th International IEEE EMBS Conference on Neural Engineering, NER 2019., 8717110, International IEEE/EMBS Conference on Neural Engineering, NER, vol. 2019-March, IEEE Computer Society, pp. 287-290, 9th International IEEE EMBS Conference on Neural Engineering, NER 2019, San Francisco, United States, 3/20/19. https://doi.org/10.1109/NER.2019.8717110

Reach-to-grasp motions: Towards a dynamic classification approach for upper-limp prosthesis. / Batzianoulis, I.; Simon, A. M.; Hargrove, L. et al.
9th International IEEE EMBS Conference on Neural Engineering, NER 2019. IEEE Computer Society, 2019. p. 287-290 8717110 (International IEEE/EMBS Conference on Neural Engineering, NER; Vol. 2019-March).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - Reach-to-grasp motions

T2 - 9th International IEEE EMBS Conference on Neural Engineering, NER 2019

AU - Batzianoulis, I.

AU - Simon, A. M.

AU - Hargrove, L.

AU - Billard, A.

PY - 2019/5/16

Y1 - 2019/5/16

N2 - During reach-to-grasp motions, the Electromyo-graphic (EMG) activity of the arm varies depending on motion stage. The variability of the EMG signals results in low classification accuracy during the reaching phase, delaying the activation of the prosthesis. To increase the efficiency of the pattern-recognition system, we investigate the muscle activity of four individuals with below-elbow amputation performing reach-to-grasp motions and segment the arm-motion into three phases with respect to the extension of the arm. Furthermore, we model the dynamic muscle contractions of each class with Gaussian distributions over the different phases and the overall motion. We quantify of the overlap among the classes with the Hellinger distance and notice larger values and, thus, smaller overlaps among the classes with the segmentation to motion phases. A Linear Discriminant Analysis classifier with phase segmentation affects positively the classification accuracy by 6 - 10% on average.

AB - During reach-to-grasp motions, the Electromyo-graphic (EMG) activity of the arm varies depending on motion stage. The variability of the EMG signals results in low classification accuracy during the reaching phase, delaying the activation of the prosthesis. To increase the efficiency of the pattern-recognition system, we investigate the muscle activity of four individuals with below-elbow amputation performing reach-to-grasp motions and segment the arm-motion into three phases with respect to the extension of the arm. Furthermore, we model the dynamic muscle contractions of each class with Gaussian distributions over the different phases and the overall motion. We quantify of the overlap among the classes with the Hellinger distance and notice larger values and, thus, smaller overlaps among the classes with the segmentation to motion phases. A Linear Discriminant Analysis classifier with phase segmentation affects positively the classification accuracy by 6 - 10% on average.

UR - http://www.scopus.com/inward/record.url?scp=85066738058&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85066738058&partnerID=8YFLogxK

U2 - 10.1109/NER.2019.8717110

DO - 10.1109/NER.2019.8717110

M3 - Conference contribution

AN - SCOPUS:85066738058

T3 - International IEEE/EMBS Conference on Neural Engineering, NER

SP - 287

EP - 290

BT - 9th International IEEE EMBS Conference on Neural Engineering, NER 2019

PB - IEEE Computer Society

Y2 - 20 March 2019 through 23 March 2019

ER -

Reach-to-grasp motions: Towards a dynamic classification approach for upper-limp prosthesis

Abstract

Publication series

Conference

Funding

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this