Prediction of five-class finger flexion using ECoG signals

Brain Computer Interface (BCI) is one of the clinical applications that might restore communication to people with severe motor disabilities. Recording and analysis of electrophysiological brain signals is the base of BCI research and development. Electrocorticography (ECoG) is an invasive record to brain signals from electrode grids on the surface of the brain. ECoG signal makes possible localization of the source of neural signals with respect to certain brain functions due to its high spatial resolution. This study is a step towards exploring the usability of ECoG signals as a BCI input technique and a multidimensional BCI control. Signal processing and classification were validated to predict kinematic parameters for five-class finger flexion. The signal is provided by ECoG dataset from BCI competition IV. For features extraction we used shift invariant wavelet decomposition and multi-taper frequency spectrum. Multilayer perceptron and pace regression were used for classification. Results show that the predicted finger movement is highly correlated with movement states.