Classification for myoelectric control system 1 using adaptive resonance theory 2 3

15 This research proposes an exploratory study of a simple, accurate, and computationally 16 efficient movement classification technique for prosthetic hand application. Surface 17 myoelectric signals were acquired from the four muscles, namely, flexor carpi ulnaris, 18 extensor carpi radialis, biceps brachii, and triceps brachii, of four normal-limb subjects. The 19 signals were segmented, and the features were extracted with a new combined time-domain 20 feature extraction method. Fuzzy C-means clustering method and scatter plot were used to 21 evaluate the performance of the proposed multi-feature versus Hudgins’ multi-feature. The 22 movements were classified with a hybrid Adaptive Resonance Theory-based neural network. 23 Comparative results indicate that the proposed hybrid classifier not only has good 24 classification accuracy (89.09%) but also a significantly improved computation time. 25