A parallel classification strategy to simultaneous control elbow, wrist, and hand movements.
Multi-DoFs control
Pattern recognition
Prosthetic control
Real-time and offline performance
Upper limb
Journal
Journal of neuroengineering and rehabilitation
ISSN: 1743-0003
Titre abrégé: J Neuroeng Rehabil
Pays: England
ID NLM: 101232233
Informations de publication
Date de publication:
28 01 2022
28 01 2022
Historique:
received:
12
08
2021
accepted:
03
01
2022
entrez:
29
1
2022
pubmed:
30
1
2022
medline:
22
3
2022
Statut:
epublish
Résumé
In the field of myoelectric control systems, pattern recognition (PR) algorithms have become always more interesting for predicting complex electromyography patterns involving movements with more than 2 Degrees of Freedom (DoFs). The majority of classification strategies, used for the prosthetic control, are based on single, hierarchical and parallel linear discriminant analysis (LDA) classifiers able to discriminate up to 19 wrist/hand gestures (in the 3-DoFs case), considering both combined and discrete motions. However, these strategies were introduced to simultaneously classify only 2 DoFs and their use is limited by the lack of online performance measures. This study introduces a novel classification strategy based on the Logistic Regression (LR) algorithm with regularization parameter to provide simultaneous classification of 3 DoFs motion classes. The parallel PR-based strategy was tested on 15 healthy subjects, by using only six surface EMG sensors. Twenty-seven discrete and complex elbow, hand and wrist motions were classified by keeping the number of electromyographic (EMG) electrodes to a bare minimum and the classification error rate under 10 %. To this purpose, the parallel classification strategy was implemented by using three classifiers one for each DoF: the "Elbow classifier", the "Wrist classifier", and the "Hand classifier" provided the simultaneous control of the elbow, hand, and wrist joints, respectively. Both the offline and real-time performance metrics were evaluated and compared with the LDA parallel classification results. The real-time recognition results were statistically better with the LR classifier with respect to the LDA classifier, for all motion classes (elbow, hand and wrist). In this paper, a novel parallel PR-based strategy was proposed for classifying up to 3 DoFs: three joint classifiers were employed simultaneously for classifying 27 motion classes related to the elbow, wrist, and hand and promising results were obtained.
Sections du résumé
BACKGROUND
In the field of myoelectric control systems, pattern recognition (PR) algorithms have become always more interesting for predicting complex electromyography patterns involving movements with more than 2 Degrees of Freedom (DoFs). The majority of classification strategies, used for the prosthetic control, are based on single, hierarchical and parallel linear discriminant analysis (LDA) classifiers able to discriminate up to 19 wrist/hand gestures (in the 3-DoFs case), considering both combined and discrete motions. However, these strategies were introduced to simultaneously classify only 2 DoFs and their use is limited by the lack of online performance measures. This study introduces a novel classification strategy based on the Logistic Regression (LR) algorithm with regularization parameter to provide simultaneous classification of 3 DoFs motion classes.
METHODS
The parallel PR-based strategy was tested on 15 healthy subjects, by using only six surface EMG sensors. Twenty-seven discrete and complex elbow, hand and wrist motions were classified by keeping the number of electromyographic (EMG) electrodes to a bare minimum and the classification error rate under 10 %. To this purpose, the parallel classification strategy was implemented by using three classifiers one for each DoF: the "Elbow classifier", the "Wrist classifier", and the "Hand classifier" provided the simultaneous control of the elbow, hand, and wrist joints, respectively.
RESULTS
Both the offline and real-time performance metrics were evaluated and compared with the LDA parallel classification results. The real-time recognition results were statistically better with the LR classifier with respect to the LDA classifier, for all motion classes (elbow, hand and wrist).
CONCLUSIONS
In this paper, a novel parallel PR-based strategy was proposed for classifying up to 3 DoFs: three joint classifiers were employed simultaneously for classifying 27 motion classes related to the elbow, wrist, and hand and promising results were obtained.
Identifiants
pubmed: 35090512
doi: 10.1186/s12984-022-00982-z
pii: 10.1186/s12984-022-00982-z
pmc: PMC8796482
doi:
Types de publication
Journal Article
Research Support, Non-U.S. Gov't
Langues
eng
Sous-ensembles de citation
IM
Pagination
10Informations de copyright
© 2022. The Author(s).
Références
Front Neurorobot. 2016 Sep 07;10:9
pubmed: 27656140
J Neuroeng Rehabil. 2017 Aug 14;14(1):82
pubmed: 28807038
IEEE Trans Neural Syst Rehabil Eng. 2011 Apr;19(2):186-92
pubmed: 21193383
Acta Orthop Scand. 1973;44(4):389-409
pubmed: 4771275
J Neural Eng. 2018 Dec;15(6):066026
pubmed: 30229745
IEEE Trans Biomed Eng. 2003 Jul;50(7):848-54
pubmed: 12848352
J Rehabil Res Dev. 2011;48(6):643-59
pubmed: 21938652
IEEE Trans Neural Syst Rehabil Eng. 2010 Apr;18(2):185-92
pubmed: 20071269
IEEE Trans Biomed Eng. 2013 May;60(5):1250-8
pubmed: 23247839
IEEE Trans Biomed Eng. 2008 Aug;55(8):1956-65
pubmed: 18632358
Crit Rev Biomed Eng. 2002;30(4-6):459-85
pubmed: 12739757
Sensors (Basel). 2013 Sep 17;13(9):12431-66
pubmed: 24048337
Annu Int Conf IEEE Eng Med Biol Soc. 2013;2013:4223-6
pubmed: 24110664
J Neuroeng Rehabil. 2014 May 30;11:91
pubmed: 24886664
IEEE Trans Neural Syst Rehabil Eng. 2010 Aug;18(4):424-32
pubmed: 20378481
Bull Prosthet Res. 1970 Spring;10(13):135-64
pubmed: 5521906
Sensors (Basel). 2021 Mar 10;21(6):
pubmed: 33802231
Annu Int Conf IEEE Eng Med Biol Soc. 2013;2013:1599-602
pubmed: 24110008
Semin Plast Surg. 2015 Feb;29(1):62-72
pubmed: 25685105
J Neurosci Methods. 2020 Jan 15;330:108462
pubmed: 31711883
Australas Phys Eng Sci Med. 2018 Jun;41(2):549-559
pubmed: 29744809
J Orthop Res. 2004 Jan;22(1):208-13
pubmed: 14656682
JAMA. 2009 Feb 11;301(6):619-28
pubmed: 19211469
CBE Life Sci Educ. 2013 Fall;12(3):345-51
pubmed: 24006382
J Neuroeng Rehabil. 2014 Jan 10;11:5
pubmed: 24410948
Annu Int Conf IEEE Eng Med Biol Soc. 2007;2007:6134-7
pubmed: 18003415
Front Psychol. 2013 Nov 26;4:863
pubmed: 24324449
IEEE Trans Neural Syst Rehabil Eng. 2014 Jul;22(4):727-34
pubmed: 24760931
IEEE Trans Neural Syst Rehabil Eng. 2014 Jul;22(4):756-64
pubmed: 24710833
Sci Rep. 2017 Oct 23;7(1):13840
pubmed: 29062019
Sensors (Basel). 2019 Jul 18;19(14):
pubmed: 31323888