Control of Hand Prosthesis Using Fusion of Biosignals and Information from Prosthesis Sensors
In this chapter we present an advanced method of recognition of patient’s intention to move the multi-articulated prosthetic hand during grasping and manipulation of objects in a skillful manner. The proposed method is based on a 2-level multiclassier system (MCS) with base classifiers dedicated to particular types of biosignals: electromyography (EMG) and mechanomyography (MMG) signals, and with combining mechanism using a dynamic ensemble selection scheme and probabilistic competence function. To improve the precision and reliability of prosthesis control, the feedback signal derived from the prosthesis sensors is used. We present two original concepts of using such a signal. In the 1st method, the feedback signal is treated as a source of information about a correct class of hand movement and competence functions of base classifiers are dynamically tuned according to this information. In the 2nd one, classification procedure is organized into multistage process based on a decision tree scheme and consequently, feedback signal indicating an interior node of a tree allows us to narrow down the set of classes. The performance of MCS with feedback signal were experimentally tested on real datasets. The obtained results show that developed methodology can be practically applied to design a control system for dexterous bioprosthetic hand.
KeywordsBase Classifier Feedback Signal Short Time Fourier Transform Rest Position Correct Class
This work was financed from the National Science Center resources in 2012–2014 years as a research project No ST6/06168, and from Wroclaw University of Technology as a statutory project.
- 2.De Luca, C.: Electromyography. In: Webster, J.G. (ed.) Encyclopedia of Medical Devices and Instrumentation, pp. 98–109. Wiley, New Jersey (2006)Google Scholar
- 3.Englehart, K.: Signal representation for classification of the transient myoelectric signal. Ph.D. Thesis, University of New Brunswick, Fredericton, New Brunswick (1998)Google Scholar
- 5.Kurzynski, M., Wolczowski, A.: Dynamic selection of classifier ensemble applied to the recognition of EMG signal for the control of bioprosthetic hand. In: Proceedings of 11th International Conference on Control, Automation and Systems, Seoul, pp. 175–182 (2011)Google Scholar
- 6.Kurzynski, M., Wolczowski, A., Tito, A.: Control of bioprosthetic hand based on EMG and MMG signals recognition using multiclassifier system with feedback from the prosthesis sensors. In: Proceedings of 2nd International Conference on Systems and Control, Marrakesch, pp. 127–135 (2012)Google Scholar
- 7.Kurzynski, M., Wolczowski, A.: Classification of EMG signals in a system for training of bioprosthetic hand control in one side handless human. In: Proceedings of International Conference on Electrical Engineering and Computer Science EECS12, Szanghai, pp. 566–576 (2012)Google Scholar
- 8.Nishikawa, D.: Studies on Electromyogram to Motion Classifier. Ph.D. Thesis, Graduate School of Engineering, Hokkaido University, Sapporo (2001)Google Scholar
- 9.Orizio, C.: Muscle sound: basis for the introduction of a mechanomyographic signal in muscle studies. Crit. Rev. Biomed. Eng. 21, 201–243 (1993)Google Scholar
- 10.Wolczowski, A.: Smart hand: the concept of sensor based control. In: Proceedings of 7th IEEE International Symposium MMAR, Miedzyzdroje, pp. 783–790 (2001)Google Scholar
- 11.Wolczowski, A., Kurzynski, M.: Control of Artificial Hand Via Recognition of EMG Signals, LNCS, vol. 3337. Springer, New York (2004)Google Scholar
- 13.Wolczowski, A., Krysztoforski, K.: Artificial hand control via EMG signal classification—experimental investigation of algorithms. In: Tchon, K., Warszawa, W.K.L. (ed.) Progress in Robotics, pp. 97–122 (2008)Google Scholar
- 15.Wolczowski, A., Kurzynski, M., Zaplotny, P.: Concept of a system for training of biopros-thetic hand control in one side handless humans using virtual reality and visual and sensory biofeedback. Med. Inf. Technol. 18, 85–91 (2012)Google Scholar