Neural Independent Component Analysis of Experimental Electro-Myographic Data from non-Invasive Surface Sensors

In this paper, we report on a biomedical engineering research project we are carrying out in coopertion with Duke University, Durham, NC, USA. The neural schemes implementing Indipendent Component Analysis (ICA) are used in order to process a vector of experimental electro-myographic data coming from surface sensors. Previous works on the subject used directly sEMG data from muscles thus generating cross-talk. Also, sEMG data are combined with fMRI data with the aim of correlating muscular to cortical activity.