Self-Correcting Pattern Recognition System of Surface EMG Signals for Upper Limb Prosthesis Control

Published on Apr 1, 2014in IEEE Transactions on Biomedical Engineering4.491
· DOI :10.1109/TBME.2013.2296274
Sebastian Amsüss5
Estimated H-index: 5
(GAU: University of Göttingen),
Peter M. Goebel5
Estimated H-index: 5
+ 3 AuthorsDario Farina80
Estimated H-index: 80
(GAU: University of Göttingen)
Pattern recognition methods for classifying user motion intent based on surface electromyography developed by research groups in well-controlled laboratory conditions are not yet clinically viable for upper limb prosthesis control, due to their limited robustness in users' real-life situations. To address this problem, a novel postprocessing algorithm, aiming to detect and remove misclassifications of a pattern recognition system of forearm and hand motions, is proposed. Using the maximum likelihood calculated by a classifier and the mean global muscle activity of the forearm, an artificial neural network was trained to detect potentially erroneous classification decisions. This system was compared to four previously proposed classification postprocessing methods, in both able-bodied and amputee subjects. Various nonstationarities were included in the experimental protocol to account for challenges posed in real-life settings, such as different contraction levels, static and dynamic motion phases, and effects induced by day-to-day transfers, such as electrode shifts, impedance changes, and psychometric user variability. The improvement in classification accuracy with respect to the unprocessed classifier ranged from 4.8% to 31.6%, depending on the scenarios investigated. The system significantly reduced misclassifications to wrong active classes and is thus a promising approach for improving the robustness of hand prosthesis controllability.
  • References (36)
  • Citations (88)
📖 Papers frequently viewed together
968 Citations
1,193 Citations
424 Citations
78% of Scinapse members use related papers. After signing in, all features are FREE.
#1Nan GeH-Index: 3
#2Peter M. GoebelH-Index: 5
Last. Dario FarinaH-Index: 80
view all 6 authors...
The robustness of myo-electric prosthesis usage is largely influenced by user performance, where psychological factors (i.e. cognitive-skills, motor-skills and psychological status - such as motivation, will, and stress) play a prominent role. These factors become more important the more degrees of freedom (DOF) a multifunctional prosthesis provides. Despite the large amount of research efforts during the past decades on developing robust control and feedback methods, there has been limited atte...
5 CitationsSource
#1Lukai LiuH-Index: 5
#2Pu LiuH-Index: 6
Last. Kevin Englehart (UNB: University of New Brunswick)H-Index: 24
view all 5 authors...
Time and frequency domain features of the surface electromyogram (EMG) signal acquired from multiple channels have frequently been investigated for use in controlling upper-limb prostheses. A common control method is EMG-based motion classification. We propose the use of EMG signal whitening as a preprocessing step in EMG-based motion classification. Whitening decorrelates the EMG signal and has been shown to be advantageous in other EMG applications including EMG amplitude estimation and EMG-fo...
33 CitationsSource
Jul 1, 2013 in EMBC (International Conference of the IEEE Engineering in Medicine and Biology Society)
#1Sebastian Amsüss (GAU: University of Göttingen)H-Index: 5
#2Liliana P. ParedesH-Index: 6
Last. Dario Farina (GAU: University of Göttingen)H-Index: 80
view all 6 authors...
Long-term functioning of a hand prosthesis is crucial for its acceptance by patients with upper limb deficit. In this study the reliability over days of the performance of pattern classification approaches based on surface electromyography (sEMG) signal for the control of upper limb prostheses was investigated. Recordings of sEMG from the forearm muscles were obtained across five consecutive days from five healthy subjects. It was demonstrated that the classification performance decreased monoto...
25 CitationsSource
#1Erik Scheme (UNB: University of New Brunswick)H-Index: 20
#2B. Hudgins (UNB: University of New Brunswick)H-Index: 19
Last. Kevin Englehart (UNB: University of New Brunswick)H-Index: 24
view all 3 authors...
This study describes a novel myoelectric control scheme that is capable of motion rejection. As an extension of the commonly used linear discriminant analysis (LDA), this system generates a confidence score for each decision, providing the ability to reject those with a score below a selected threshold. The thresholds are class-specific and affect only the rejection characteristics of the associated class. Furthermore, because the rejection stage is implemented using the outputs of the LDA, the ...
45 CitationsSource
#1Xinpu Chen (SJTU: Shanghai Jiao Tong University)H-Index: 4
#2Dingguo Zhang (SJTU: Shanghai Jiao Tong University)H-Index: 21
Last. Xiangyang Zhu (SJTU: Shanghai Jiao Tong University)H-Index: 26
view all 3 authors...
Background The nonstationary property of electromyography (EMG) signals usually makes the pattern recognition (PR) based methods ineffective after some time in practical application for multinational prosthesis. The conventional EMG PR, which is accomplished in two separate steps: training and testing, ignores the mismatch between training and testing conditions and often discards the useful information in testing dataset.
64 CitationsSource
#1Erik Scheme (UNB: University of New Brunswick)H-Index: 20
#2Kevin EnglehartH-Index: 24
The performance of pattern recognition based myoelectric control has seen significant interest in the research community for many years. Due to a recent surge in the development of dexterous prosthetic devices, determining the clinical viability of multifunction myoelectric control has become paramount. Several factors contribute to differences between offline classification accuracy and clinical usability, but the overriding theme is that the variability of the elicited patterns increases great...
48 CitationsSource
#1Tatiana Tommasi (Idiap Research Institute)H-Index: 20
#2Francesco Orabona (Toyota)H-Index: 25
Last. Barbara Caputo (Idiap Research Institute)H-Index: 44
view all 4 authors...
At the time of this writing, the main means of control for polyarticulated self-powered hand prostheses is surface electromyography (sEMG). In the clinical setting, data collected from two electrodes are used to guide the hand movements selecting among a finite number of postures. Machine learning has been applied in the past to the sEMG signal (not in the clinical setting) with interesting results, which provide more insight on how these data could be used to improve prosthetic functionality. R...
51 CitationsSource
#1Yanjuan Geng (CAS: Chinese Academy of Sciences)H-Index: 10
#2Ping Zhou (Rehabilitation Institute of Chicago)H-Index: 22
Last. Guanglin Li (CAS: Chinese Academy of Sciences)H-Index: 26
view all 3 authors...
Background Electromyography (EMG) pattern-recognition based control strategies for multifunctional myoelectric prosthesis systems have been studied commonly in a controlled laboratory setting. Before these myoelectric prosthesis systems are clinically viable, it will be necessary to assess the effect of some disparities between the ideal laboratory setting and practical use on the control performance. One important obstacle is the impact of arm position variation that causes the changes of EMG p...
88 CitationsSource
#1Anders Lyngvi Fougner (NTNU: Norwegian University of Science and Technology)H-Index: 7
#2Øyvind Stavdahl (NTNU: Norwegian University of Science and Technology)H-Index: 23
Last. Philip A. Parker (UNB: University of New Brunswick)H-Index: 31
view all 5 authors...
The recent introduction of novel multifunction hands as well as new control paradigms increase the demand for advanced prosthetic control systems. In this context, an unambiguous terminology and a good understanding of the nature of the control problem is important for efficient research and communication concerning the subject. Thus, one purpose of this paper is to suggest an unambiguous taxonomy, applicable to control systems for upper limb prostheses and also to prostheses in general. A funct...
208 CitationsSource
Aug 1, 2012 in EMBC (International Conference of the IEEE Engineering in Medicine and Biology Society)
#1Caitlin L. Chicoine (Rehabilitation Institute of Chicago)H-Index: 2
#2Ann M. Simon (Rehabilitation Institute of Chicago)H-Index: 19
Last. Levi J. Hargrove (Rehabilitation Institute of Chicago)H-Index: 35
view all 3 authors...
Pattern recognition can provide intuitive control of myoelectric prostheses. Currently, screen-guided training (SGT), in which individuals perform specific muscle contractions in sync with prompts displayed on a screen, is the common method of collecting the electromyography (EMG) data necessary to train a pattern recognition classifier. Prosthesis-guided training (PGT) is a new data collection method that requires no additional hardware and allows the individuals to keep their focus on the pros...
14 CitationsSource
Cited By88
#1Joseph L. Betthauser (Johns Hopkins University)H-Index: 5
#2John T. Krall (Johns Hopkins University)H-Index: 1
Last. Nitish V. Thakor (Johns Hopkins University)H-Index: 66
view all 7 authors...
Prediction of movement intentions from electromyographic (EMG) signals is typically performed with a pattern recognition approach, wherein a short dataframe of raw EMG is compressed into an instantaneous feature-encoding that is meaningful for classification. However, EMG signals are time-varying, implying that a frame-wise approach may not sufficiently incorporate temporal context into predictions, leading to erratic and unstable prediction behavior. Goal: We demonstrate that sequential predict...
#1Evan Campbell (UNB: University of New Brunswick)H-Index: 3
#2Angkoon Phinyomark (UNB: University of New Brunswick)H-Index: 23
Last. Erik Scheme (UNB: University of New Brunswick)H-Index: 24
view all 3 authors...
This manuscript presents a hybrid study of a comprehensive review and a systematic(research) analysis. Myoelectric control is the cornerstone ofmany assistive technologies used in clinicalpractice, 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 in...
4 CitationsSource
#1Pandia Rajan Jeyaraj (Mepco Schlenk Engineering College)H-Index: 2
#2Edward Rajan Samuel Nadar (Mepco Schlenk Engineering College)H-Index: 3
In this research paper we present designing and evaluating the electrocardiography (ECG) and Myoelectric signal (EMG) pattern recognition methods based on the adaptive machine learning. For this theoretical model to describe how the Boundary Misclassification Risk (BMR) changes along parameters including, the adaptive learning times, the adaptive learning frequencies, the generalization ability of the predictive model, and the ratio of samples without supervised information during the adaptive l...
#1Agamemnon Krasoulis (Edin.: University of Edinburgh)H-Index: 6
#2Sethu Vijayakumar (Newcastle University)H-Index: 34
Last. Kianoush Nazarpour (Newcastle University)H-Index: 18
view all 3 authors...
In the field of upper-limb myoelectric prosthesis control, the use of statistical and machine learning methods has been long proposed as a means of enabling intuitive grip selection and actuation. Recently, this paradigm has found its way toward commercial adoption. Machine learning-based prosthesis control typically relies on the use of a large number of electrodes. Here, we propose an end-to-end strategy for multi-grip, classification-based prosthesis control using only two sensors, comprising...
1 CitationsSource
#1Ali Ameri (Shahid Beheshti University of Medical Sciences and Health Services)H-Index: 7
#2Mohammad Ali Akhaee (UT: University of Tehran)H-Index: 12
Last. Kevin Englehart (UNB: University of New Brunswick)H-Index: 24
view all 4 authors...
An important barrier to commercialization of pattern recognition myoelectric control of prostheses is the lack of robustness to confounding factors such as electrode shift, skin impedance variations, and learning effects. To overcome this challenge, a novel supervised adaptation approach based on transfer learning (TL) with convolutional neural networks (CNNs) is proposed which requires only a short training session (a few seconds for each class) to recalibrate the system. TL is proposed as a so...
#1Mojisola Grace Asogbon (CAS: Chinese Academy of Sciences)H-Index: 3
#2Oluwarotimi Williams Samuel (CAS: Chinese Academy of Sciences)H-Index: 14
Last. Guanglin Li (CAS: Chinese Academy of Sciences)H-Index: 26
view all 9 authors...
Abstract Background and Objective Mobility of subject (MoS) and muscle contraction force variation (MCFV) have been shown to individually degrade the performance of multiple degrees of freedom electromyogram (EMG) pattern recognition (PR) based prostheses control systems. Though these factors (MoS-MCFV) co-exist simultaneously in the practical use the prosthesis, their combined impact on PR-based system has rarely been studied especially in the context of amputees who are the target users of the...
1 CitationsSource
#1Bin Yu (USTC: University of Science and Technology of China)H-Index: 1
#2Xu Zhang (USTC: University of Science and Technology of China)H-Index: 14
Last. Xun Chen (USTC: University of Science and Technology of China)H-Index: 19
view all 5 authors...
Pattern-recognition-based myoelectric control systems are not yet widely available due to their limited robustness in real-life situations. Some postprocessing methods were introduced to improve the robustness in previous studies, but there is lack of investigation into movement transition phases. This article presents a novel postprocessing method based on movement pattern transition (MPT) detection. An image-based index is used to quantify the similarity of adjacent feature matrices from high-...
This paper aims to propose an innovative adaptive control method for lower-limb rehabilitation robots.,Despite carrying out various studies on the subject of rehabilitation robots, the flexibility and stability of the closed-loop control system is still a challenging problem. In the proposed method, surface electromyography (sEMG) and human force-based dual closed-loop control strategy is designed to adaptively control the rehabilitation robots. A motion analysis of human lower limbs is performe...
#1Ikram Brahim (University of Paris)
#2Islame Dhibou (University of Paris)
Last. Amine Nait-Ali (University of Paris)H-Index: 11
view all 5 authors...
In this chapter, a case study using a specific wearable Multi-Channel EMG device will be considered. In particular, eight EMG channels will be used through Myo Armband system. The purpose is to deploy a verification biometric system using EMG signals corresponding to hand gestures. More specifically, the idea behind this concept is the capacity to generate a digital signature for each specific hand gesture.