A Novel Prosthetic Hand Classification Approach Using LibSVM Classification Method

Abstract Support Vector Machines (LibSVM) is widely implemented machine learning method. In this investigation, LibSVM is used to classify the feature vector of the extracted forearm surface electromyographic (sEMG) signal. The performance of fourteen features and different five feature combinations were tested in order to achieve the highest classification accuracy of different eleven hand motions. In addition, genetic algorithm (GA) was implemented to select the optimal parameters of LibSVM classifier. The result shows that root mean square (RMS) is a robust feature outperforms other features using LibSVM and achieved 93% average classification accuracy. From our experimental result, the highest classification accuracy was achieved by sample entropy (SampEnt), root mean square (RMS), myopulse percentage rate (MYOP) and difference absolute standard deviation value (DASDV) feature combination with 97% using the proposed LibSVM classification method. The experimental results reveal that our pattern recognition method performs better than the existing classification algorithm SVM with 5% increment in the classification accuracy based on best selected feature combination.

[1]  R.N. Scott,et al.  A new strategy for multifunction myoelectric control , 1993, IEEE Transactions on Biomedical Engineering.

[2]  J. Richman,et al.  Physiological time-series analysis using approximate entropy and sample entropy. , 2000, American journal of physiology. Heart and circulatory physiology.

[3]  Minoru Fukumi,et al.  Wrist EMG pattern recognition system by neural networks and genetic algorithms , 2004 .

[4]  Congli Mei,et al.  Pattern Recognition of Eight Hand Motions Using Feature Extraction of Forearm EMG Signal , 2014 .

[5]  Ali Lashgari,et al.  Hard-rock LHD cost estimation using single and multiple regressions based on principal component analysis , 2012 .

[6]  K. Thangavel,et al.  A guided ant colony algorithm for detection of microcalcification clusters , 2012 .

[7]  Anders Lyngvi Fougner,et al.  Proportional Myoelectric Control of a Multifunction Upper-limb Prosthesis , 2007 .

[8]  M. Osman Tokhi,et al.  A fuzzy clustering neural network architecture for multifunction upper-limb prosthesis , 2003, IEEE Transactions on Biomedical Engineering.

[9]  Hessam Mirgolbabaei,et al.  A novel principal component analysis-based acceleration scheme for LES-ODT: An a priori study , 2012 .

[10]  Hong-Bo Xie,et al.  Ant colony optimization-based feature selection method for surface electromyography signals classification , 2012, Comput. Biol. Medicine.

[11]  Sazali Yaacob,et al.  A comparative study of wavelet families for classification of wrist motions , 2012, Comput. Electr. Eng..

[12]  Kevin B. Englehart,et al.  A wavelet-based continuous classification scheme for multifunction myoelectric control , 2001, IEEE Transactions on Biomedical Engineering.

[13]  E. Lamounier,et al.  A VIRTUAL PROSTHESIS CONTROL BASED ON NEURAL NETWORKS FOR EMG PARTTERN CLASSIFICATION , 2005 .

[14]  Christos Koukouvinos,et al.  Classification methods and ROC analysis for outcome prediction of patients following injuries , 2012 .

[15]  Chi-Woong Mun,et al.  Comparison of k-nearest neighbor, quadratic discriminant and linear discriminant analysis in classification of electromyogram signals based on the wrist-motion directions , 2011 .

[16]  Chandrakar Kamath.  ECG classification using morphological features derived from symbolic dynamics , 2012 .

[17]  Kevin B. Englehart,et al.  A robust, real-time control scheme for multifunction myoelectric control , 2003, IEEE Transactions on Biomedical Engineering.

[18]  Mohammad Hassan Moradi,et al.  Evaluation of the forearm EMG signal features for the control of a prosthetic hand. , 2003, Physiological measurement.

[19]  Angkoon Phinyomark,et al.  EMG feature evaluation for improving myoelectric pattern recognition robustness , 2013, Expert Syst. Appl..

[20]  Tanu Sharma,et al.  A novel feature extraction for robust EMG pattern recognition , 2016, Journal of medical engineering & technology.

[21]  Dennis C. Tkach,et al.  Study of stability of time-domain features for electromyographic pattern recognition , 2010, Journal of NeuroEngineering and Rehabilitation.

[22]  Hakil Kim,et al.  Novel and efficient pedestrian detection using bidirectional PCA , 2013, Pattern Recognit..

[23]  Chih-Jen Lin,et al.  LIBSVM: A library for support vector machines , 2011, TIST.

[24]  Bruce C. Wheeler,et al.  EMG feature evaluation for movement control of upper extremity prostheses , 1995 .

[25]  Ivan Prebil,et al.  Multivariate and multiscale monitoring of large-size low-speed bearings using Ensemble Empirical Mode Decomposition method combined with Principal Component Analysis , 2010 .

[26]  Pornchai Phukpattaranont,et al.  Feature reduction and selection for EMG signal classification , 2012, Expert Syst. Appl..

[27]  Pornchai Phukpattaranont,et al.  A Novel Feature Extraction for Robust EMG Pattern Recognition , 2009, ArXiv.

[28]  Ismail A. Ismail,et al.  An efficient offline signature identification method based on Fourier Descriptor and chain codes , 2011 .