A Novel Pattern Classification Method for Multivariate EMG Signals Using Neural Network

• Published in: Advances in Natural Computation pp. 165 - 174
• Main Authors: Bu, Nan, Arita, Jun, Tsuji, Toshio
• Format: Book Chapter Conference Proceeding
• Language: English; Japanese
• Published: Berlin, Heidelberg Springer Berlin Heidelberg 2005; Springer
• Series: Lecture Notes in Computer Science
• Subjects: Applied sciences; Artificial intelligence; Cervical Muscle; Computer science; control theory; systems; Exact sciences and technology; Feature Extraction Technique; Gaussian Mixture Model; Principal Component Analysis; Probabilistic Neural Network; Mixed distribution; Dimensionality; Probabilistic approach; Extraction process; Pattern recognition; Neural network; Modeling; Pattern classification; Phonetics; Feature extraction; Electromyography; Feasibility; Phoneme; Pattern extraction; Principal component analysis
• ISBN: 9783540283256; 3540283250; 3540283234; 9783540283232
• ISSN: 0302-9743; 1611-3349
• DOI: 10.1007/11539117_26

Feature extraction is an important issue in electromyography (EMG) pattern classification, where feature sets of high dimensionality are always used. This paper proposes a novel classification method to deal with high-dimensional EMG patterns, using a probabilistic neural network, a reduced-dimensional log-linearized Gaussian mixture network (RD-LLGMN) [1]. Since RD-LLGMN merges feature extraction and pattern classification processes into its structure, lower-dimensional feature set consistent with classification purposes can be extracted, so that, better classification performance is possible. To verify feasibility of the proposed method, phoneme classification experiments were conducted using frequency features of EMG signals measured from mimetic and cervical muscles. Filter banks are used to extract frequency features, and dimensionality of the features grows significantly when we increase resolution of frequency. In these experiments, the proposed method achieved considerably high classification rates, and outperformed traditional methods that are based on principle component analysis (PCA).