2017
DOI: 10.1155/2017/5090454
|View full text |Cite
|
Sign up to set email alerts
|

Improving the Robustness of Real-Time Myoelectric Pattern Recognition against Arm Position Changes in Transradial Amputees

Abstract: Previous studies have showed that arm position variations would significantly degrade the classification performance of myoelectric pattern-recognition-based prosthetic control, and the cascade classifier (CC) and multiposition classifier (MPC) have been proposed to minimize such degradation in offline scenarios. However, it remains unknown whether these proposed approaches could also perform well in the clinical use of a multifunctional prosthesis control. In this study, the online effect of arm position vari… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
2
1

Citation Types

2
54
0

Year Published

2018
2018
2023
2023

Publication Types

Select...
6
1

Relationship

2
5

Authors

Journals

citations
Cited by 40 publications
(56 citation statements)
references
References 44 publications
2
54
0
Order By: Relevance
“…This leads to lower classification accuracy that results in a slower reaction of the prosthesis. This outcome is aligned with the findings of [38], where a classifier that takes into account different arm positions outperformed a single-position classifier. The benefits of a dynamic training protocol are also shown in [40].…”
Section: Discussionsupporting
confidence: 84%
See 1 more Smart Citation
“…This leads to lower classification accuracy that results in a slower reaction of the prosthesis. This outcome is aligned with the findings of [38], where a classifier that takes into account different arm positions outperformed a single-position classifier. The benefits of a dynamic training protocol are also shown in [40].…”
Section: Discussionsupporting
confidence: 84%
“…Although the classification results for our proposed approach are comparable with previous studies [31,36,37], it is different for two main reasons. Previous studies examine the classification performance of different hand gestures, including wrist motion, hand open/close, a small number of grasp types (2-4) and in some cases the resting condition [38][39][40]. Whereas, we focus only on grasping gestures with different finger configurations.…”
Section: Discussionmentioning
confidence: 99%
“…The most commonly reported limb position experimental protocol in the literature is one that uses static limb positions [21,22,94,[113][114][115][116]119,[121][122][123][126][127][128][129][130][131][132][133][134][135][136][137][138]. Among the articles surveyed, these make up 50% of all experiments.…”
Section: Static Limb Positionmentioning
confidence: 99%
“…2 EMG signals that provide information about the muscle activation patterns have been widely applied in various fields including sports training, neural rehabilitation, and game control in virtual augmented reality. [7][8][9] In clinical settings, the acquisition of ECG, EMG, and EEG signals often require the use of conductive gel, which may make it uncomfortable for patients and therefore limit the application of physiological signal monitoring. 3 In addition, EEG signals provide a means of diagnosing different kind of intracranial lesions (such as stroke, 4 brain tumors, 5 encephalitis, metabolic brain lesions, and also detecting the motor intent of individuals for the control of rehabilitation robots 6 ).…”
Section: Introductionmentioning
confidence: 99%
“…In addition, EEG recordings could reflect the state of sleep and by monitoring the EEG waveform during sleep, sleep quality and sleep stages can be determined. [7][8][9] In clinical settings, the acquisition of ECG, EMG, and EEG signals often require the use of conductive gel, which may make it uncomfortable for patients and therefore limit the application of physiological signal monitoring. 10 During long-term physiological signal monitoring, the conductive gel will gradually dry out and the signal quality will eventually degrade, 11 making it the most important obstacle for conventional long-term health care monitoring.…”
Section: Introductionmentioning
confidence: 99%