Classification of muscle groups related to neuropathy disease by modeling EMG signals

Oezsert M., Yildirim T., Baslo B.

IEEE 15th Signal Processing and Communications Applications Conference, Eskişehir, Türkiye, 11 - 13 Haziran 2007, ss.945-946 identifier identifier

  • Cilt numarası:
  • Doi Numarası: 10.1109/siu.2007.4298712
  • Basıldığı Şehir: Eskişehir
  • Basıldığı Ülke: Türkiye
  • Sayfa Sayıları: ss.945-946


Purpose of this work is to classify three different muscle types. For this purpose, the Electromyogram (EMG) signals were recorded from Biceps, Frontallis, Abductor Pollisis brevis muscles. For the modelling of EMG signals, Autoregressive models used and Autoregressive coefficients used to train and test several Artificial Neural Networks (ANNs). The results of experiments show that Radial Basis Function neural network has 93,3% accuracy to classificate the muscles. After this classifying stage the next step will be the diagnosis of Neuropathy dissease which is defined as the communication damage of nerves between organs and tissue.