Limb Position Tolerant Pattern Recognition for Myoelectric Prosthesis Control with Adaptive Sparse Representations From Extreme Learning

Betthauser, Joseph L.; Hunt, Christopher L.; Osborn, Luke; Masters, Matthew; Lévay, György; Kaliki, Rahul R.; Thakor, Nitish V.

doi:10.1109/tbme.2017.2719400

Cited by 84 publications

(69 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The first method of minimizing the effect of limb position is the exploration of feature extraction, dimensionality reduction, and classification algorithms that are inherently less susceptible to this effect [23,94,116,130,135,140,[150][151][152]. In an exploratory study conducted by Liu et al [116], repeatability metrics between positions were computed for various commonly used feature sets.…”

Section: Robust Algorithmsmentioning

confidence: 99%

“…The application of SRC yielded significantly better performance across all combinations of training and testing positions (p < 0.001). Additionally, the use of extreme learning machines (ELM) in adaptive sparse representation classification (EASRC) greatly reduced computational burden of SRC while maintaining stability under untrained limb positions during an online experiment [150].…”

Section: Robust Algorithmsmentioning

confidence: 99%

“…However, by instead allowing dynamic motions that span static positions, or free motion of the arm in the 3D space, a greater breadth of limb positions and forearm orientations can be measured in less time [22,156,157]. A second consideration of this method involves the possibility that inter-position variability, if in excess, may ambiguate decision boundaries, increasing inter-class error in a single stage classifier [150].…”

Section: Training Strategiesmentioning

confidence: 99%

See 2 more Smart Citations

Current Trends and Confounding Factors in Myoelectric Control: Limb Position and Contraction Intensity

Campbell

Phinyomark

Scheme

2020

Sensors

View full text Add to dashboard Cite

This manuscript presents a hybrid study of a comprehensive review and a systematic (research) analysis. Myoelectric control is the cornerstone of many assistive technologies used in clinical practice, such as prosthetics and orthoses, and human-computer interaction, such as virtual reality control. Although the classification accuracy of such devices exceeds 90% in a controlled laboratory setting, myoelectric devices still face challenges in robustness to variability of daily living conditions. The intrinsic physiological mechanisms limiting practical implementations of myoelectric devices were explored: the limb position effect and the contraction intensity effect. The degradation of electromyography (EMG) pattern recognition in the presence of these factors was demonstrated on six datasets, where classification performance was 13% and 20% lower than the controlled setting for the limb position and contraction intensity effect, respectively. The experimental designs of limb position and contraction intensity literature were surveyed. Current state-of-the-art training strategies and robust algorithms for both effects were compiled and presented. Recommendations for future limb position effect studies include: the collection protocol providing exemplars of at least 6 positions (four limb positions and three forearm orientations), three-dimensional space experimental designs, transfer learning approaches, and multi-modal sensor configurations. Recommendations for future contraction intensity effect studies include: the collection of dynamic contractions, nonlinear complexity features, and proportional control.

show abstract

Section: Robust Algorithmsmentioning

confidence: 99%

Section: Robust Algorithmsmentioning

confidence: 99%

Section: Training Strategiesmentioning

confidence: 99%

See 1 more Smart Citation

Current Trends and Confounding Factors in Myoelectric Control: Limb Position and Contraction Intensity

Campbell

Phinyomark

Scheme

2020

Sensors

View full text Add to dashboard Cite

show abstract

“…A complicating factor of sEMG in the context of natural myocontrol is the limb position effect [6,7,8,9,10,11], namely the change in signals depending on the position and orientation of the limb (body posture). The application of PR by its very nature requires training data that sufficiently captures these variations and it has therefore become bestpractice to perform the acquisition procedure in multiple positions [6,8,12]. Although this strategy seems to be effective in addressing the limb position effect, it comes at the cost of a considerably longer and more tiring acquisition protocol.…”

Section: Introductionmentioning

confidence: 99%

Natural Myocontrol in a Realistic Setting: a Comparison Between Static and Dynamic Data Acquisition

Gigli

Gijsberts

Castellini

2019

2019 IEEE 16th International Conference on Rehabilitation Robotics (ICORR)

View full text Add to dashboard Cite

Natural myocontrol employs pattern recognition to allow users to control a robotic limb intuitively using their own voluntary muscular activations. The reliability of myocontrol strongly depends on the signals initially collected from the users, which must appropriately capture the variability encountered later on during operation. Since myoelectric signals can vary based on the position and orientation of the limb, it has become best practice to gather data in multiple body postures. We hereby concentrate on this acquisition protocol and investigate the relative merits of collecting data either statically or dynamically. In the static case, data for a desired hand configuration is collected while the users keep their hand still in certain positions, whereas in the dynamic case, data is collected while users move their limbs, passing through the required positions with a roughly constant velocity.Fourteen able-bodied subjects were asked to naturally control two dexterous hand prostheses mounted on splints, performing a set of complex, realistic bimanual activities of daily living. We could not find any significant difference between the protocols in terms of the total execution times, although the dynamic data acquisition was faster and less tiring. This would indicate that dynamic data acquisition should be preferred over the static one.

show abstract

“…Comparing the classifiers performance, Adewuyi et al (2016) found for non-amputees and partial-hand amputees that LDA and ANN perform better than the quadratic discriminant analysis. Betthauser et al (2018) developed a robust sparsity-based adaptive classification method to get a classification system which is appreciably less sensitive to signal deviations between training and testing. When they tested it on eight able-bodied and two transradial amputee subjects with eight electrodes pairs regularly spaced around the proximal forearm, it was found that their approach significantly outperformed other movement classification methods.…”

Section: Introductionmentioning

confidence: 99%

Biceps Brachii Muscle Synergy and Target Reaching in a Virtual Environment

Mathieu

2019

Front. Neurorobot.

View full text Add to dashboard Cite

A muscular synergy is a theory suggesting that the central nervous system uses few commands to activate a group of muscles to produce a given movement. Here, we investigate how a muscle synergy extracted from a single muscle can be at the origin of different signals which could facilitate the control of modern upper limb myoelectric prostheses with many degrees of freedom. Five pairs of surface electrodes were positioned across the biceps of 12 normal subjects and electromyographic (EMG) signals were collected while their upper limbs were in eight different static postures. Those signals were used to move, within a virtual cube, a small red sphere toward different targets. With three muscular synergies extracted from the five EMG signals, a classifier was trained to identify which synergy pattern was associated with a given static posture. Later, when a posture was recognized, the result was a displacement of a red sphere toward a corner of a virtual cube presented on a computer screen. The axes of the cube were assigned to the shoulder, elbow and wrist joint while each of its the corners was associated with a static posture. The goal for subjects was to reach, one at a time, the four targets positioned at different locations and heights in the virtual cube with different sequences of postures. The results of 12 normal subjects indicate that with the muscular synergies of the biceps brachii, it was possible, but not easy for an untrained person, to reach a target on each trial. Thus, as a proof of concept, we show that features of the biceps muscular synergy have the potential to facilitate the control of upper limb myoelectric prostheses. To our knowledge, this has never been shown before.

show abstract

Limb Position Tolerant Pattern Recognition for Myoelectric Prosthesis Control with Adaptive Sparse Representations From Extreme Learning

Abstract: This method of prosthesis control has the potential to deliver real-world clinical benefits to amputees: better condition-tolerant performance, reduced training burden in terms of frequency and duration, and increased adoption of myoelectric prostheses.

Cited by 84 publications

References 37 publications

Current Trends and Confounding Factors in Myoelectric Control: Limb Position and Contraction Intensity

Current Trends and Confounding Factors in Myoelectric Control: Limb Position and Contraction Intensity

Natural Myocontrol in a Realistic Setting: a Comparison Between Static and Dynamic Data Acquisition

Biceps Brachii Muscle Synergy and Target Reaching in a Virtual Environment

Contact Info

Product

Resources

About