{"title":"Fuzzy Wavelet Packet based Feature Extraction Method for Multifunction Myoelectric Control","authors":"Rami N. Khushaba, Adel Al-Jumaily","volume":13,"journal":"International Journal of Mechanical and Mechatronics Engineering","pagesStart":126,"pagesEnd":135,"ISSN":"1307-6892","URL":"https:\/\/publications.waset.org\/pdf\/13053","abstract":"The myoelectric signal (MES) is one of the Biosignals\r\nutilized in helping humans to control equipments. Recent approaches\r\nin MES classification to control prosthetic devices employing pattern\r\nrecognition techniques revealed two problems, first, the classification\r\nperformance of the system starts degrading when the number of\r\nmotion classes to be classified increases, second, in order to solve the\r\nfirst problem, additional complicated methods were utilized which\r\nincrease the computational cost of a multifunction myoelectric\r\ncontrol system. In an effort to solve these problems and to achieve a\r\nfeasible design for real time implementation with high overall\r\naccuracy, this paper presents a new method for feature extraction in\r\nMES recognition systems. The method works by extracting features\r\nusing Wavelet Packet Transform (WPT) applied on the MES from\r\nmultiple channels, and then employs Fuzzy c-means (FCM)\r\nalgorithm to generate a measure that judges on features suitability for\r\nclassification. Finally, Principle Component Analysis (PCA) is\r\nutilized to reduce the size of the data before computing the\r\nclassification accuracy with a multilayer perceptron neural network.\r\nThe proposed system produces powerful classification results (99%\r\naccuracy) by using only a small portion of the original feature set.","references":"[1] K. Englehart, B. Hudgin, and P.A. Parker, \"A wavelet-based\r\ncontinuous classification scheme for multifunction myoelectric\r\ncontrol,\" IEEE Transactions on Biomedical Engineering, vol. 48, pp.\r\n302 - 311, 2001.\r\n[2] C. J. De Luca, \"Physiology and mathematics of myoelectric signals,\"\r\nIEEE Transactions on Biomedical Engineering, vol. BME-26, pp. 313-\r\n325, 1979.\r\n[3] R. Merletti and P. Parker, \"Electromyography physiology, Engineering,\r\nand noninvasive applications,\" IEEE Press Engineering in Medicine\r\nand Biology Society., 2004.\r\n[4] J. Hunt, \"A 3-degree-of-freedom myoelectric control suitable for easy\r\nimplementation in hardware,\" in Electrical and Computer Engineering,\r\nvol. Master of Science. Fredericton, NB Canada: University of\r\nNewBrunswick, 1998.\r\n[5] B. Hudgins, P. Parker, and R. N. Scott, \"A new strategy for\r\nmultifunction myoelectric control,\" IEEE Transactions on Biomedical\r\nEngineering, vol. 40, pp. 82-94, 1993.\r\n[6] C. M. Lighty, P. H. Chappelly, B. Hudgins, and K. Englehart,\r\n\"Intelligent multifunction myoelectric control of hand prostheses,\"\r\nJournal of Medical Engineering & Technology, vol. 26, pp. 139- 146,\r\n2002.\r\n[7] S. Leowinata, \"A new strategy for multifunction myoelectric control\r\nusing an array of surface electrodes,\" in Electrical and Computer\r\nEngineering, vol. Master: University of New Brunswick, 2000.\r\n[8] M. Vuskovic and D. Sijiang, \"Classification of prehensile EMG\r\npatterns with simplified fuzzy ARTMAP networks,\" Proceedings of the\r\nInternational Joint Conference on Neural Networks, IJCNN '02. , vol.\r\n3, pp. 2539-2544, 2002.\r\n[9] J. J. Im, D. H. Rho, Y. J. Jeon, N. B. Lee, and J. I. Chung, \"Extraction\r\nof parameters from EMG signals for the biofeedback electrical\r\nstimulation,\" presented at Proceedings of the Second Joint\r\nEMBS\/BMES Conference, 2002.\r\n[10] S-P. Lee, S-H. Park, J-S. Kim, and I.-J. Kim, \"EMG pattern recognition\r\nbased on evidence accumulation for prosthesis control,\" 18th Annual\r\nInternational Conference of the IEEE Engineering in Medicine and\r\nBiology Society, Amsterdam, The Netherlands, pp. 1481-1483, 1996.\r\n[11] Sang-Hui Park and S.-P. Lee, \"EMG Pattern Recognition Based on\r\nArtificial Intelligence Techniques,\" IEEE Transactions on\r\nRehabilitation Engineering, vol. 6, pp. 400-405, 1998.\r\n[12] L. Seok-Pil, P. Sang-Hui, K. Jeong-Seop, and K. Ig-Jae, \"EMG pattern\r\nrecognition based on evidence accumulation for prosthesis control,\"\r\n18th Annual International Conference of the IEEE Engineering in\r\nMedicine and Biology Society, Amsterdam, vol. 4, pp. 1481-1483 vol.4,\r\n1996.\r\n[13] Jun-Uk Chu, Inhyuk Moon, Shin-Ki Kim, and M.-S. Mun., \"Control of\r\nmultifunction myoelectric hand using a real-time EMG pattern\r\nrecognition,\" IEEE\/RSJ International Conference on Intelligent Robots\r\nand Systems (IROS 2005). pp. 3511 - 3516, 2005.\r\n[14] Y. H. Kim, I. J. Shim, and G. T. Park, \"A method of controlling\r\nhousehold electrical appliance by hand motion in LonWorks,\"\r\nProceedings of the 41st SICE Annual Conference SICE. , vol. 5, pp.\r\n2849-2854, 2002.\r\n[15] B. Hannaford and S. Lehman, \"Short time fourier analysis of the\r\nelectromyogram: fast movements and constant contraction,\" IEEE\r\nTransactions on Biomedical Engineering, vol. BME-33, pp. 1173-\r\n1181, 1986.\r\n[16] S. Karlsson, Y. Jun, and M. Akay, \"Time-frequency analysis of\r\nmyoelectric signals during dynamic contractions: a comparative study,\"\r\nIEEE Transactions on Biomedical Engineering, vol. 47, pp. 228-238,\r\n2000.\r\n[17] M. Zecca, S. Micera, M. C. Carrozza, and P. Dario, \"Control of\r\nmultifunctional prosthetic hands by processing the electromyographic\r\nsignal,\" Critical Review in Biomedical Engineering, vol. 30, pp. 459-\r\n485, 2002.\r\n[18] K. Englehart, B. Hudgins, P. A. Parker, and M. Stevenson, \"Improving\r\nmyoelectric signal classification using wavelet packets and principal\r\ncomponents analysis,\" IEEE Engineering in Medicine and Biology\r\nSociety, Atlanta, vol. 1, pp. 569, 1999.\r\n[19] J.-U. Chu, I. Moon, and M.-S. Mun, \"A real-time EMG pattern\r\nrecognition based on linear-nonlinear feature projection for\r\nmultifunction myoelectric hand,\" Proceedings of the IEEE 9th\r\nInternational Conference on Rehabilitation Robotics, pp. 295-298,\r\n2005.\r\n[20] J. U. Chu, I. Moon, and M. S. Mun, \"A real-time EMG pattern\r\nrecognition system based on linear-nonlinear feature projection for a\r\nmultifunction myoelectric hand,\" IEEE Transactions on Biomedical\r\nEngineering, vol. 53, pp. 2232-2239, 2006.\r\n[21] L. Hargrove, K. Englehart, and B. Hudgins, \"A comparison of surface\r\nand intramuscluar myoelectric signal classification,\" IEEE\r\nTransactions on Biomedical Engineering, Accepted for future\r\npublication.\r\n[22] R. R. Coifman, Y. Meyer, and M. V. Wickerhauser, \"Wavelet analysis\r\nand signal processing,\" in Wavelets and Their Applications, M. B.\r\nRuskai, Ed. Boston: Jones and Bartlett, 1992.\r\n[23] K. Englehart and B. Hudgins, \"A robust, real-time control scheme for\r\nmultifunction myoelectric control,\" Biomedical Engineering, IEEE\r\nTransactions on, vol. 50, pp. 848-854, 2003.\r\n[24] L. Deqiang, W. Pedrycz, and N. J. Pizzi, \"Fuzzy wavelet packet based\r\nfeature extraction method and its application to biomedical signal\r\nclassification,\" IEEE Transactions on Biomedical Engineering, vol. 52,\r\npp. 1132-1139, 2005.\r\n[25] K. Englehart, \"Signal representation for classification of the transient\r\nmyoelectric signal \" in Electrical and Computer Engineering\r\nDepartment., vol. PhD Dissertation: University of New Brunswick,\r\n1998.\r\n[26] I. Drummond and S. Sandri, \"A clustering-based fuzzy classifier,\" in\r\nArtificial Intelligence Research and Development, B. Lopez, Ed.: IOS\r\nPress, 2005.\r\n[27] Y. KO\u251c\u00e7Y\u2500\u2591\u2500\u00d7\u2500\u2591T and M. KOR\u251c\u00a3REK, \"EMG signal classif\u2500\u2592cation using\r\nwavelet transform and fuzzy clustering algorithms,\" presented at Proc.\r\nof ELECO'2003, Bursa, Turkey, 2003.\r\n[28] B. Karlik, M. Osman Tokhi, and M. Alci, \"A fuzzy clustering neural\r\nnetwork architecture for multifunction upper-limb prosthesis,\" IEEE\r\nTransactions on Biomedical Engineering, vol. 50, pp. 1255-1261,\r\n2003.\r\n[29] A. D. Boca and D. C. Park, \"Myoelectric signal recognition using fuzzy\r\nclustering and artificial neural networks in real time,\" IEEE\r\nInternational Conference on Neural Networks, vol. 5, pp. 3098-3103\r\n1994.\r\n[30] M. M. Trivedi and J. C. Bezdeck, \"Low-level segmentation of aerial\r\nimages with fuzzy clustering,\" IEEE Transactions on Systems, Man and\r\nCybernetics, vol. SMC-16, pp. 589- 598, 1986.\r\n[31] R. Boostani and M. H. Moradi, \"Evaluation of the forearm EMG signal\r\nfeatures for the control of a prosthetic hand,\" Physiological\r\nMeasurement, vol. 24, pp. 309-319, 2003.","publisher":"World Academy of Science, Engineering and Technology","index":"Open Science Index 13, 2008"}