Real-time identification of μ wave with wavelet neural networks

In the rehabilitation of paralyzed patients, the (f) under bar unctional (e) under bar lectrical (s) under bar timulation (FES) or prostheses often adopted in clinical practiceOne of the key issues in these new technologies is the source for generating control commands. The (b) under bar rain-(c) under bar omputer (i) under bar nterface (BCI)creates an alternative pathway from the braipotentials. In this investigation, we construct real-time system to percept the voluntary movement of right thumb as a basic study of BCI. We combine the wavelet transformationand neural network as Wavelet Neural Network (WNN) to identify the attempt of voluntary thumb movement. Three types of classification methods: realtime classification without network update, real-time classification with update and conventional power spectral analyses are compared, and it was found that the WNN with off-line retraining shows better successful rate up to 80%.