Surface EMG based handgrip force predictions using gene expression programming

Zhong, Yang; Chen, Yumiao; Tang, Zhichuan; Wang, Jianping

doi:10.1016/j.neucom.2016.05.038

Cited by 35 publications

(17 citation statements)

References 64 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The onset of one drawing trial is designated as the onset of feature extraction. As a time domain feature, the Root Mean Square (RMS) represents the characteristic of the amplitude change of EMG signals on the time dimension, which can nondestructively measure the state of muscle activity in real-time (Yang et al, 2016 ). The RMS is also widely accepted (Shrirao et al, 2009 ; Tang et al, 2015 ).…”

Section: Methodsmentioning

confidence: 99%

“…First of all, on account of the characteristics of simplicity, high efficiency, and functional complexity, GEP combines the advantages of both GAs and GP, while overcoming some of their limitations, which offers great potentiality to solve complex modeling and optimization problems (Zhou et al, 2003 ). Moreover, after the training process, GEP can produce simple explicit formulas with high accuracy (Landeras et al, 2012 ) and reduce the number of sEMG features (Yang et al, 2016 ). In addition, our proposed GEP model showed promise for recognizing sketching based on sEMG signals (Yang and Chen, 2016 ).…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

A Novel Hybrid Model for Drawing Trace Reconstruction from Multichannel Surface Electromyographic Activity

Chen

Yang

2017

Front. Neurosci.

Self Cite

View full text Add to dashboard Cite

Recently, several researchers have considered the problem of reconstruction of handwriting and other meaningful arm and hand movements from surface electromyography (sEMG). Although much progress has been made, several practical limitations may still affect the clinical applicability of sEMG-based techniques. In this paper, a novel three-step hybrid model of coordinate state transition, sEMG feature extraction and gene expression programming (GEP) prediction is proposed for reconstructing drawing traces of 12 basic one-stroke shapes from multichannel surface electromyography. Using a specially designed coordinate data acquisition system, we recorded the coordinate data of drawing traces collected in accordance with the time series while 7-channel EMG signals were recorded. As a widely-used time domain feature, Root Mean Square (RMS) was extracted with the analysis window. The preliminary reconstruction models can be established by GEP. Then, the original drawing traces can be approximated by a constructed prediction model. Applying the three-step hybrid model, we were able to convert seven channels of EMG activity recorded from the arm muscles into smooth reconstructions of drawing traces. The hybrid model can yield a mean accuracy of 74% in within-group design (one set of prediction models for all shapes) and 86% in between-group design (one separate set of prediction models for each shape), averaged for the reconstructed x and y coordinates. It can be concluded that it is feasible for the proposed three-step hybrid model to improve the reconstruction ability of drawing traces from sEMG.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

A Novel Hybrid Model for Drawing Trace Reconstruction from Multichannel Surface Electromyographic Activity

Chen

Yang

2017

Front. Neurosci.

Self Cite

View full text Add to dashboard Cite

show abstract

“…The k-nearest neighbor (k-NN) and support vector machine (SVM) methods were applied to classify the normal and compensatory movement patterns (trunk lean forward, trunk rotation and shoulder elevation) during three basic seated reaching tasks, including side-to-side, back-and-forth, and up-and-down reaching. Furthermore, the surface electromyography (sEMG) signals, which can reflect the degree of activity of the muscles [28], was used to provide a more detailed description of compensatory movements and verify the classification results.…”

Section: Introductionmentioning

confidence: 99%

Detecting compensatory movements of stroke survivors using pressure distribution data and machine learning algorithms

Cai

Zhang

et al. 2019

J NeuroEngineering Rehabil

View full text Add to dashboard Cite

Background: Compensatory movements are commonly employed by stroke survivors during seated reaching and may have negative effects on their long-term recovery. Detecting compensation is useful for coaching the patient to reduce compensatory trunk movements and improving the motor function of the paretic arm. Sensor-based and camera-based systems have been developed to detect compensatory movements, but they still have some limitations, such as causing object obstructions, requiring complex setups and raising privacy concerns. To overcome these drawbacks, this paper proposes a compensatory movement detection system based on pressure distribution data and is unobtrusive, simple and practical. Machine learning algorithms were applied to classify compensatory movements automatically. Therefore, the purpose of this study was to develop and test a pressure distribution-based system for the automatic detection of compensation movements of stroke survivors using machine learning algorithms. Methods: Eight stroke survivors performed three types of reaching tasks (back-and-forth, side-to-side, and up-anddown reaching tasks) with both the healthy side and the affected side. The pressure distribution data were recorded, and five features were extracted for classification. The k-nearest neighbor (k-NN) and support vector machine (SVM) algorithms were applied to detect and categorize the compensatory movements. The surface electromyography (sEMG) signals of nine trunk muscles were acquired to provide a detailed description and explanation of compensatory movements. Results: Cross-validation yielded high classification accuracies (F1-score>0.95) for both the k-NN and SVM classifiers in detecting compensation movements during all the reaching tasks. In detail, an excellent performance was achieved in discriminating between compensation and noncompensation (NC) movements, with an average F1score of 0.993. For the multiclass classification of compensatory movement patterns, an average F1-score of 0.981 was achieved in recognizing the NC, trunk lean-forward (TLF), trunk rotation (TR) and shoulder elevation (SE) movements. Conclusions: Good classification performance in detecting and categorizing compensatory movements validated the feasibility of the proposed pressure distribution-based system. Reliable classification accuracy achieved by the machine learning algorithms indicated the potential to monitor compensation movements automatically by using the pressure distribution-based system when stroke survivors perform seated reaching tasks.

show abstract

“…It can also be found that the proposed GEP classifier presented the highest accuracy and robust when compared with the BPNN classifier and the ENN classifier. One of the advantages of the GEP classifier is that it can produce simple explicit formulations (Landeras et al, 2012 ; Yang et al, 2016 ), which gives a better understanding of the derived relationship between the sEMG signals and one-stroke sketching shapes and makes the model suitable for application in real time. Although the ENN classifier achieved higher accuracy rate than the BPNN classifier in our previous work (Chen et al, 2015 ), it performed worst in this study.…”

Section: Resultsmentioning

confidence: 99%

Surface EMG-based Sketching Recognition Using Two Analysis Windows and Gene Expression Programming

Yang

Chen

2016

Front. Neurosci.

Self Cite

View full text Add to dashboard Cite

Sketching is one of the most important processes in the conceptual stage of design. Previous studies have relied largely on the analyses of sketching process and outcomes; whereas surface electromyographic (sEMG) signals associated with sketching have received little attention. In this study, we propose a method in which 11 basic one-stroke sketching shapes are identified from the sEMG signals generated by the forearm and upper arm muscles from 4 subjects. Time domain features such as integrated electromyography, root mean square and mean absolute value were extracted with analysis windows of two length conditions for pattern recognition. After reducing data dimensionality using principal component analysis, the shapes were classified using Gene Expression Programming (GEP). The performance of the GEP classifier was compared to the Back Propagation neural network (BPNN) and the Elman neural network (ENN). Feature extraction with the short analysis window (250 ms with a 250 ms increment) improved the recognition rate by around 6.4% averagely compared with the long analysis window (2500 ms with a 2500 ms increment). The average recognition rate for the eleven basic one-stroke sketching patterns achieved by the GEP classifier was 96.26% in the training set and 95.62% in the test set, which was superior to the performance of the BPNN and ENN classifiers. The results show that the GEP classifier is able to perform well with either length of the analysis window. Thus, the proposed GEP model show promise for recognizing sketching based on sEMG signals.

show abstract

Surface EMG based handgrip force predictions using gene expression programming

Cited by 35 publications

References 64 publications

A Novel Hybrid Model for Drawing Trace Reconstruction from Multichannel Surface Electromyographic Activity

A Novel Hybrid Model for Drawing Trace Reconstruction from Multichannel Surface Electromyographic Activity

Detecting compensatory movements of stroke survivors using pressure distribution data and machine learning algorithms

Surface EMG-based Sketching Recognition Using Two Analysis Windows and Gene Expression Programming

Contact Info

Product

Resources

About