Distance and mutual information methods for EMG feature and channel subset selection for classification of hand movements

Abstract Different approaches have been proposed to select features and channels for pattern recognition classification of myoelectric upper-limb prostheses. The goal of this work is to use deterministic methods to select the feature-channels pairs that best classify the hand postures at different limb positions. Two selection methods were tried. One is a distance-based feature selection (DFSS) that determines a separability index using the Mahalanobis distance between classes. The second method is a correlation-based feature selection (CFSS) that measures the amount of mutual information between features and classes. To evaluate the performance of these selection methods, EMG data from 10 able-bodied subjects were acquired when performing 5 hand postures at 9 different arm positions and 10 time-domain and frequency-domain features were extracted. Classification accuracy using both methods was always higher than including all the features and channels and showed slight improvement over classification using the state-of-art TD features when evaluated against limb variation. The CFSS method always used less feature-channel pairs compared to the DFSS method. Using both methods, selection of channels placed on the posterior side of the forearm was significantly higher than anterior side. Such methods could be used as fast screening filters to select features and channels that best classify different hand postures at different arm positions.

[1]  S H Park,et al.  EMG pattern recognition based on artificial intelligence techniques. , 1998, IEEE transactions on rehabilitation engineering : a publication of the IEEE Engineering in Medicine and Biology Society.

[2]  Ilja Kuzborskij,et al.  On the challenge of classifying 52 hand movements from surface electromyography , 2012, 2012 Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[3]  Erik Scheme,et al.  Training Strategies for Mitigating the Effect of Proportional Control on Classification in Pattern Recognition–Based Myoelectric Control , 2013, Journal of prosthetics and orthotics : JPO.

[4]  Jaime Valls Miró,et al.  Towards limb position invariant myoelectric pattern recognition using time-dependent spectral features , 2014, Neural Networks.

[5]  K. Englehart,et al.  Resolving the Limb Position Effect in Myoelectric Pattern Recognition , 2011, IEEE Transactions on Neural Systems and Rehabilitation Engineering.

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

[7]  C. Cipriani,et al.  The Effects of Weight and Inertia of the Prosthesis on the Sensitivity of Electromyographic Pattern Recognition in Relax State , 2012 .

[8]  Christian Cipriani,et al.  Feature and Channel Selection Using Correlation Based Method for Hand Posture Classification in Multiple Arm Positions , 2014 .

[9]  P. Mahalanobis On the generalized distance in statistics , 1936 .

[10]  Keinosuke Fukunaga,et al.  Introduction to Statistical Pattern Recognition , 1972 .

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

[12]  Shuxiang Guo,et al.  Comparison of sEMG-Based Feature Extraction and Motion Classification Methods for Upper-Limb Movement , 2015, Sensors.

[13]  Henri Maître,et al.  On the relevance of linear discriminative features , 2010, Inf. Sci..

[14]  Fuhui Long,et al.  Feature selection based on mutual information criteria of max-dependency, max-relevance, and min-redundancy , 2003, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[15]  Chih-Jen Lin,et al.  Asymptotic Behaviors of Support Vector Machines with Gaussian Kernel , 2003, Neural Computation.

[16]  B. Hudgins,et al.  The effect of electrode displacements on pattern recognition based myoelectric control , 2006, 2006 International Conference of the IEEE Engineering in Medicine and Biology Society.

[17]  Yong Deng,et al.  A novel feature selection method based on CFS in cancer recognition , 2012, 2012 IEEE 6th International Conference on Systems Biology (ISB).

[18]  Frederick E. Croxton,et al.  Applied General Statistics. , 1940 .

[19]  Ashley N. Johnson,et al.  Dual-task motor performance with a tongue-operated assistive technology compared with hand operations , 2012, Journal of NeuroEngineering and Rehabilitation.

[20]  Guanglin Li,et al.  EMG feature assessment for myoelectric pattern recognition and channel selection: a study with incomplete spinal cord injury. , 2014, Medical engineering & physics.

[21]  F. Finley,et al.  Myocoder studies of multiple myopotential response. , 1967, Archives of physical medicine and rehabilitation.

[22]  Weidong Yang,et al.  Erratum to “Biodistribution and SPECT Imaging Study of 99mTc Labeling NGR Peptide in Nude Mice Bearing Human HepG2 Hepatoma” , 2014, BioMed Research International.

[23]  Guanglin Li,et al.  Toward attenuating the impact of arm positions on electromyography pattern-recognition based motion classification in transradial amputees , 2012, Journal of NeuroEngineering and Rehabilitation.

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

[25]  F. K. Lam,et al.  Fuzzy EMG classification for prosthesis control. , 2000, IEEE transactions on rehabilitation engineering : a publication of the IEEE Engineering in Medicine and Biology Society.

[26]  Gang Wang,et al.  The Analysis of Hand Movement Distinction Based on Relative Frequency Band Energy Method , 2014, BioMed research international.

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

[28]  Huosheng Hu,et al.  Myoelectric control systems - A survey , 2007, Biomed. Signal Process. Control..

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

[30]  Zhaojie Ju,et al.  Surface EMG Based Hand Manipulation Identification Via Nonlinear Feature Extraction and Classification , 2013, IEEE Sensors Journal.

[31]  David G. Stork,et al.  Pattern Classification , 1973 .

[32]  Levi J. Hargrove,et al.  A training strategy to reduce classification degradation due to electrode displacements in pattern recognition based myoelectric control , 2008, Biomed. Signal Process. Control..

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

[34]  Huosheng Hu,et al.  Feature-channel subset selection for optimising myoelectric human-machine interface design , 2013 .

[35]  D Graupe,et al.  Multifunctional prosthesis and orthosis control via microcomputer identification of temporal pattern differences in single-site myoelectric signals. , 1982, Journal of biomedical engineering.

[36]  J. Roitman,et al.  ACSM's Resource Manual for Guidelines for Exercise Testing and Prescription , 1998 .

[37]  Max Ortiz-Catalan,et al.  BioPatRec: A modular research platform for the control of artificial limbs based on pattern recognition algorithms , 2013, Source Code for Biology and Medicine.

[38]  Christian Antfolk,et al.  Decoding of individuated finger movements using surface EMG and input optimization applying a genetic algorithm , 2011, 2011 Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[39]  Niki Pissinou,et al.  Correlation-Based Feature Selection for Intrusion Detection Design , 2007, MILCOM 2007 - IEEE Military Communications Conference.

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

[41]  Erik Scheme,et al.  Electromyogram pattern recognition for control of powered upper-limb prostheses: state of the art and challenges for clinical use. , 2011, Journal of rehabilitation research and development.

[42]  Nathan E. Bunderson,et al.  Quantification of Feature Space Changes With Experience During Electromyogram Pattern Recognition Control , 2012, IEEE Transactions on Neural Systems and Rehabilitation Engineering.

[43]  A. Al-Jumaily,et al.  Channel and Feature Selection in Multifunction Myoelectric Control , 2007, 2007 29th Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[44]  Lloyd A. Smith,et al.  Feature Selection for Machine Learning: Comparing a Correlation-Based Filter Approach to the Wrapper , 1999, FLAIRS.

[45]  Finley Fr,et al.  Myocoder studies of multiple myopotential response. , 1967 .